Graduate labour market update: weakened but not in meltdown

Charlie Ball, head of labour market intelligence at Jisc, considers the state of the graduate labour market in the UK, highlighting the key factors behind its noticeable weakening - before summarising the latest news and reports

First, a brief commentary. There have been a lot of stories of late suggesting that the UK graduate labour market is in some kind of unprecedented meltdown. I simply don't see it, and nor do most of the specialists in the field I speak to.

It is clear that the labour market has been weakening for a while and has got noticeably weaker in the last 12 months, and this seems largely due to three major factors.

The first and easiest is that post-COVID the UK jobs market was extremely strong from around mid-2021 to the end of 2023. I think a lot of people now commentating really missed how unusual that jobs market was - we went through a period of fewer jobseekers than vacancies, which while not unusual in the US, is very unusual in the UK - and for quite a while the labour market was substantially outperforming the rather anaemic UK economy. That was never going to last forever, and since 2024 the labour market has been much more closely tracking the UK economy.

The second is the impact of National Insurance rises in 2024 on employer costs and their willingness and ability to hire has been substantial. It is cited by employers time and time again as a key factor in reduced hiring. Related to this in particular is the budget restraints on parts of the public sector, particularly the NHS, that has frozen hiring in crucial graduate professions despite there being demand. There isn't the money to hire people, not a lack of demand for their services.

The third seems to be AI, but not in the way commonly cited. Despite a lot of rhetoric, there is little evidence of widespread replacement of jobs, even entry-level ones, by AI in the UK. That's not to say employers aren't considering it, but as the Bank of England said last month, they're largely pushing forward cautiously with AI adoption and as yet there is little real evidence of it happening right now.

What AI has done instead, and particularly from summer 2024, is had a substantial impact on recruitment. Most applicants are not very experienced in writing applications for jobs, and by and large AI does it better. It also makes it far easier and where it might take hours to put in an application, what AI allows you to do is make multiple applications with much less effort, thus removing a crucial barrier to entry. It's also pretty good at maximising your positives and minimising your negatives. The upshot is last year employers were inundated with many more applications than they were expecting, used to and, crucially, were budgeted for, all of an overall higher quality, largely functionally identical and harder to sift and this made hiring more expensive. And overall the number of good, appointable candidates didn't really change, it was just a lot harder to find them. This has had a profound impact. As British Chambers of Commerce report, a lot of businesses are not recruiting even though they acknowledge that they're short-staffed, for example.

For the last 15 to 20 years the aim has been for many employers to widen their recruitment net to make sure they didn't miss any talent. For the first time in a long time many employers are trying to reduce their applicant numbers while still trying to get the candidates that they want. That could have an impact on some of the data sources we use a lot, as if employer behaviour has shifted (and there is evidence of employers being less likely to use big open jobs boards and so on to avoid being inundated) then vacancy data might be affected (I'm using caveats here, but I think it's happening, although I think vacancy data is still tremendously useful and interesting).

At present the UK is not obviously in an economic recession with the labour market recession that would likely follow. It is possible that just as the labour market became uncoupled from the wider economy in a positive way post-COVID, it could also be uncoupled now but in the other direction and that it's behaving like a recessionary labour market outside a recession, but I don't think the wider data picture supports that (although there's some things I hear that I don't like).

We haven't had a real prolonged downturn since the 2008 recession precipitated a long period of poor labour market conditions in the UK, and I think a lot of people have forgotten - or haven't really experienced - what that's like. The good news is that it's a lot worse than what we're experiencing now. The bad news is that when we do have the next recession, and the signs are not all that encouraging, it'll be worse than what we're experiencing now.

Anyway, on with the news.

Again, the number of jobs vacancies has seen a marginal drop on the month of about 9,000, and now stands at 717,000 for the three months between July and September. The general view is emerging that the sharper falls of the earlier part of 2025 seem to have ceased for now and the market seems to be holding steadier - I'm not completely sure we won't see further falls now but it does seem that over the summer and into September the jobs market has stabilised a bit. We do seem to be running a bit below the pre-pandemic vacancy rate and it looks a bit more like the labour market around the time of Brexit in 2016/17 (which was nothing to write home about).

The ratio of unemployed people to vacancies now stands at 2.4, up from 1.7 this time last year. This does show that the market did drop off quite a bit over the last 12 months, as that's quite a big rise.

Vacancy falls are particularly pronounced in real estate, in hospitality, and in health (where we know there are a lot of hiring freezes), but it is notable that professional services (professional services vacancies are still ahead of pre-COVID levels) and IT both seem to have seen a pick up in vacancies in the last couple of months and both have more vacancies than they have had since Spring, while education vacancies are also picking up - the real pain seems to be in non-graduate jobs in the service industry.

Redundancies have ticked up though, and although hospitality has very clearly been hit the most, professional services, IT, construction and education have all also gone up (a bit) for short-term unemployment. However none of these sectors are seeing much of a rise in long-term unemployment - being unemployed for 12 months or more - over the summer, and long-term unemployment for professional services professionals seems to have been dropping over the last few months  (even, thankfully, hospitality) as vacancy levels might be down from where they were but are still not too bad.

The Office for National Statistics (ONS) have also published international comparison data which shows that what's happening in the UK is not internationally exceptional. Everyone's seeing slight deteriorations in their labour markets and ours actually looks ok compared to many (our unemployment and inactivity rates are both better than the OECD average).

I lost September a bit in these updates as I was doing a lot of events and travelling, so here's the most recent Bank of England Agent's Summary, which came out a couple of weeks ago.

The Bank reports that employment intentions are little changed over the summary and point towards further gradual reductions in employment, largely through natural attrition and not filling roles rather than redundancy. Recruitment difficulties are around normal for most roles. 2025 pay settlements remain modest and look to be coming down a little. The Bank reports that if redundancies were planned for 2025, they have largely happened already, and point to National Insurance and National Living Wage rises as a key factor.

Interestingly, they also report that businesses are looking at automation to increase productivity, and that many are considering using AI but adoption, where it is taking place, is proceeding 'cautiously'. And like a lot of other sources, the Bank notes that firms are seeing more applicant numbers for roles but not a corresponding rise in quality.

REC tell a similar story in their newest Report on Jobs, with reporting that falls in recruitment agency placements have continued, but at the slowest rate for a while. Tellingly, cost pressures seem to be important here again, with “with firms often noting that employers were hesitant to take on new workers due to weaker economic conditions and cost concerns.” London seems quite hard hit (although that's not the case in the wider labour market picture), but REC reports engineering bucking the trend with an increase in demand for workers.

The Institution of Engineering and Technology surveyed engineers in March and have recently published a report and a summary:

• 76% of engineering employers struggle to recruit for key roles, with technical and specialist sustainability skills topping the list.
• 42% of engineering employers rank innovative thinking as the most vital skill for future growth, ahead of digital and technical expertise (39%).
• 58% of engineering employers currently use AI, but only 18% use it regularly. 23% who do not currently use AI plan to do so in the future. 61% expect AI to improve productivity, and 50% expect enhanced problem solving.
• 36% say their organisation lacks the skills to decarbonise by 2050. Technical/specialist sustainability skills are both the most needed (39%) and most missing (35%). Cost is the top barrier to decarbonisation: 34% cite operating costs, and 29% cite investment costs.

As a final aside, I tried to use AI to find me the any news sources that I'd missed, but although I used some fairly decent prompts it couldn't find me anything other than this (and it told me about the ONS data several times). It did try to pass off some older reports as more recent. Alas, the magic robots still can't make my life easier yet. I'll keep trying.

Written by

Charlie Ball

Head of labour market intelligence

Jisc