David Kernohan is Deputy Editor of Wonkhe

The UK has a skills shortage, and – like everything else – it’s all the fault of universities.

Specifically, universities are enticing students into attractive courses of higher education that have nothing to do with local industry – leaving us with expensive, underemployed, graduates who work in low skilled jobs and sit around eating avocado on toast and/or demolish statues. Apprenticeships, technical qualifications, and lifelong learning are what we need to #levelup and #buildbackbetter.

Does this feel like an evidence-based economic argument to you? Or could there, perhaps, be the slightest hint of a culture war going on?

What employers think of graduates

The DfEs Employer Skills Survey (last published in November 2020) generally finds that about 80 per cent of employers feel that the HE leavers they employ have the skills they need. The most recent iteration puts this at 85 per cent in England, 82 per cent in Wales, and 88 per cent in Northern Ireland.

On the face of it, this is great news – a far better satisfaction rate than for leavers at any other leave at education. Any lingering desire we may have to write “nearly 1 in 5 employers…” style headlines should be quenched when we look at the skills employers say are missing – a lack of experience or maturity (8 per cent in England), or a poor attitude or lack of motivation (6 per cent in England). The first is simply a skills-language way of saying “young”, the second does have a whiff of avocado – we don’t, alas, get a comparator for all employees.

[Full screen]

Overall something like three in ten employers in England, Wales, and Northern Ireland recruited an education leaver in the last two to three years – one in seven (14 per cent) had recruited a university leaver. Eighty per cent of these felt that graduates recruited were prepared for the role they were recruited to – fifteen percent felt otherwise. Interestingly, preparedness correlates with the level of education – there was no positive effect level three education leavers, who would be more likely to have completed a skills-focused course.

Graduate jobs for graduate workers

The chances of a graduate (from hereon I’m looking at graduates from an undergraduate course only) working in a “high skill” role after graduation are affected by two main factors – the industry they work in and the area of the country they work in. Naturally, these two interrelate – in some industries highly skilled roles are concentrated in certain local areas.

But to start off, here’s just the impact of the industry on graduate job skill levels – from the most recent HESA Graduate Outcomes release.

[Full screen]

You can see here that graduates working in the accommodation and food service industries are more likely to be classified as in an “elementary occupation”, and that those in “wholesale and retail trade” are more likely to be in sales and customer service roles. Most graduates in traditional graduate destinations like health, education, or public administration are likely to be outside the (red) “highly skilled” category that has come to indicate a graduate job.

What’s counterintuitive about this is that one of the ways a job will move into the red area in future iterations of the Standard Occupational Classification (SOC, one of two coding frames you are about to become intimately familiar with) is literally that a lot of graduates do the job. Other ways include the use of professional registration requiring graduate entry, or a material change in the job itself to use higher order skills. As the world gradually shifts towards a “skilled” economy, more jobs will become graduate jobs and more graduates will do those jobs.

Local area industries

The other factor in all this is, as I said, location. The ONS NOMIS service has data on employment right down to very tiny areas – I’ve chosen to look at local authority districts (LTLAs) as being more likely to contain sensible industrial areas, though London boroughs are perhaps less reliable here the data is still interesting enough to leave in.

To start with, let’s introduce our other coding frame: the Standard Industrial Classification (SIC). You can get this at various levels – the first plot shows the number of employment instances (either someone is on a salary in an industry, or someone is an owner of a company or sole-trader/freelance in an industry) double digit (“division”) SIC for every LTLA in England, Wales, and Scotland. You can choose your industry of interest using the box on the bottom left – this shows the areas on the map on a colour scale (more employment, more blue) and the LTLAs ranked on the right. Find your LTLA via the highlighter on the right, and you can flip between raw numbers and percentage of all employment in that LTLA.

[Full screen]

And here’s a plot showing employment by industry in each area – you can also get to this by clicking on an LTLA on the first map or chart.

[Full screen]

Industries, areas, and graduates

Knowing where particular industries are concentrated is pretty nifty, and there’s a clear link to portfolio planning at a university – especially given the “local skills needs” headwinds we currently face. But how do the dominant industries in an area compare to what graduates actually do?

Here I’m taking an unconventional approach on graduate skills. For me, if a graduate is employed in an industry, they have already demonstrated that they have the skills required to get a job in that industry. So I’m not as concerned here with subject of study – obviously some subjects do prepare graduates for particular industries, so where a provider has done well at this we will be able to see this in the data.

First up here’s a map. You chose your SIC (we’re up to section level this time), and the area colours (blue) show employment in that SIC while the provider colours (shown in the circles as yellow) show the number of graduates from that provider that work in that industry.

[Full screen]

This, to be clear – doesn’t mean graduates are working in the local area to the university in that industry, but the large number of Robert Gordon University and University of Aberdeen graduates working in the “mining and quarrying” industry probably has something to do with Aberdeen being heavily focused on the oil and gas industry (remember I’m looking at undergraduate qualifiers here only, and both universities do have offshore focused undergraduate courses).

We can maybe see this a little easier using two dashboards. The first shows the overall employment in an area on the top (blue) and the employment among graduates on the bottom (green). Select an area at the top, and then choose a provider in that area (or all providers in that area) underneath.

[Full screen]

And the second looks at all areas (blue, top) and all providers (pink, bottom) for a given industry. There’s only a few LTLAs with an appreciably sized mining/resource extraction industry – but a whole bunch of providers have graduates working in that industry. Here providers outside of Aberdeen and London may not be meeting local needs, but UK needs are clearly being addressed.

[Full screen]

A quick search of UCAS suggests that there are very few undergraduate courses focused on offshore or oil and gas, so we’d have to conclude that some species of more general graduate skills is allowing for entry into that industry.


But do you need a degree to work on a rig? Or in transport and logistics? Or in retail? Sometimes, yes. And – as above – this is going to be more likely in future.

We don’t have data on graduates by industry and skill level (SIC and SOC) but I can offer you similar data for the wider population. To be clear, this (from Nomis) isn’t great data – there are plenty of suppressions due to low numbers and a fair amount of rounding – so you should see what is presented here as being indicative only. Also, we’re now above the top of the SIC stack (wide industry groups).

First up, a map – choose the industry at the top and the skill level at the bottom, the area colours show the number of employment instances in each area. Those high skill resource extraction jobs are concentrated in Aberdeen – but there are a few others (the category bringing in energy and water more generally) in other key locations. The graph to the side shows high (blue), medium (orange), and low (red) skilled jobs in each area for that industry – you can use the highlighters at the top to find your area.

[Full screen]

What can we say about graduate employment?

Overall, graduates are good at getting jobs. When they get jobs, employers are generally happier with them than they are with new starters with a lower level of educational attainment. Job skill level (and, frankly, salary) can be seen alongside each industry as having a local area component – where a graduate works and in what industry perhaps has more to do with the local area than with what they learn at university.

Graduates are generally seen as having good technical and interpersonal skills – which I would argue suggests that what they do learn at university makes them better employees. There is no real indication that there is a skills gap that universities could address but do not – if jobs are becoming more “graduate” in nature then it feels sensible to suggest that we will need more graduates to do them.

I can’t say whether or not the skills gap issue is a culture war issue – there’s certainly a popular culture war framing – but the idea that employers want less graduates and that graduates are not ready for work would appear to be nonsensical.

2 responses to “Do local industries need local skills? And do universities deliver?

  1. This article really covers the bases and I certainly took a lot from the all the data that backs it up.
    From my perspective (as me and a Careers Practitioner) the tact being taken by the Government is is doomed – however, as your article alludes, I’m sure it will be HE’s fault, the students, industry or just anyone else for that matter that won’t fit into this archaic approach.
    What we are talking about here is placement and replacement to industries (locally?)? This approach depends on a static composition within a labour market – which it has never been. So although it is possible to identify gaps (industries, jobs, skills) locally and nationally its hard to determine the elements within the foundation they sit upon and forecast effectively years in advance. Additionally, how can this be communicated in an accurate and consistent way. Who is accountable if a change doesn’t come to fruition, changes and responsibility to graduates who happen to graduate at the wrong time?

    The other side of this are individuals. How do they affect skills demand/satisfaction and future requirements in terms of how we consume and our relationship with work…these have always been subject to change.
    Additionally, whether we look at the insinuation of local skills demand or the courses available (or more importantly, the emerging stigma around non-STEM and non-professional courses) it reminds me of a child being expected to take over a family trade or business whether they wish to or not – almost like a trait and factor that ignores a person’s trait and factor. Personally, I also dislike the reductionist approach both to employment types and employment as the headline for going to university.

    The approach required is common ground between individuals, institutions, industry and policy makers with a balance between elements of autonomy and shared responsibilities. A well funded and understood take on life-long learning or skills for life would work if it allowed individuals the resources, finance, time, inter alia to make informed choices and commitments to change in themselves and the labour market with support from Government.

    If we are to view this as an employability project, rather than a placement and replacement model that doesn’t work alongside a budget that expects an individual to choose without genuine options and support to bind themselves for a lifetime to a discipline I propose a move to an appreciation of life-long evolution and re-evolution that should be supported and understood as a natural consequence of working in a service-sector economy.

  2. Students and graduates are occupation/career focussed, not sector focussed. Engineering companies are not full of engineers. R&D companies employ marketers and accountants; scientists are probably a minority. The labour market is messy but SOC data is useful, SIC data is not. The idea that local labour markets in the UK are very different has never been less true….London aside. In part this is because the public sector plays such a large role everywhere.

Leave a Reply