Every graduate should have the chance of a decent job

The season for creating metrics about graduate employment is clearly upon us.

This morning HESA puts forward an early sketch of a way of understanding whether graduates are in jobs with a good “design and nature”. Uniquely among the recent crop of novel metrics, I really like it.

I’m going to take you there the long way round, because in our collective haste to use salary or SOCs as a proxy for work that is worthwhile, meaningful, and fulfilling we have cut across a great deal of existing international work that urgently needs to be fed into where these ideas hit higher education regulation. But I’m fairly confident you’ll like it too.

Defining “fair work”

There’s a 2015 UN Sustainable Development Goal about “decent work for all”. Rolled in with the measures on sustainable economic growth, environmentally sustainable work (and, frankly, we need this built in to any measure of job quality used in higher education policy) employment rates, and forced labour we find target 8.5, which commits signatories to:

By 2030, achieve full and productive employment and decent work for all women and men, including for young people and persons with disabilities, and equal pay for work of equal value

Although the UN specifies a pair of broad indicators dealing with equal pay and unemployment, there’s no definition provided concerning “decent” work – though the read across to UK higher education policy debates about “highly skilled” and “graduate” jobs is tempting. Though the full employment wording will be familiar to anyone familiar with the history of UK economic policy, “decent work” is a newish area of interest.

There is a UN definition, which can be found in paragraph 7 of general comment 18 from the Committee on Economic, Social and Cultural Rights:

[Decent Work] 1). Respects the fundamental rights of the human person as well as the rights of workers in terms of conditions of work safety and remuneration; 2). Provides an income allowing workers to support themselves and their families; 3). Respect the physical and mental integrity of the worker in the exercise of his/her employment

Better – but still very general. The actual UK government commissioned Matthew Taylor to look into this whole area – his “Good Work” review urged us to think about:

work that is fair and decent, with realistic scope for development and fulfilment

Measuring fair work

Laudable. But difficult to measure. At least, that’s what the Measuring Job Quality Working Group felt. It’s very easy given existing data to measure salary or working hours – how do we measure “scope for development”? Or indeed, “fulfilment”? The Carnegie UK Trust and the RSA Future Work Centre got into the weeds of this in “Measuring Good Work”, which recommending the adoption of whole new suite of job quality metrics across seven dimensions:

1. Terms of employment (job security, minimum guaranteed hours, underemployment)
2. Pay and benefits (Actual pay, satisfaction with pay)
3. Health, safety and psychosocial well-being (Physical injury, mental health)
4. Job design and nature of work (Use of skills, control, opportunities for progression, sense of purpose)
5. Social support and cohesion (Peer support, line manager relationship)
6. Voice and representation (Trade union membership, employee information, employee involvement)
7. Work-life balance (Overemployment, paid/unpaid overtime)

If we want graduates to get “good” jobs, then these dimensions represent how we know that the jobs are good. Note that salary is on there, as are measures around working hours, but these are far from the whole story.

HESA intervenes

HESA’s intervention this morning focuses on point 4 of these recommendations. We are a long way down from the UN’s broad use of “decent work” – but arguably the way people are able to use their skills, feel a sense of purpose, and have opportunities for progression are key components of what makes work “decent”.

Graduate Outcomes includes three important statements for response within “graduate reflections on activity”:

• My current activity is meaningful
• My current activity fits with my future plans
• I am utilising what I learnt during my studies in my current activity

“Current activity” here reflects the fact that not all graduates have employment as their primary activity, but if you look at the questions regarding graduates that are in full time employment you get a reasonable proxy for the three aspects of the job design and nature of work. It’s true that Graduate Outcomes doesn’t quite fit the data requirements set out – it’s not a long standing survey, at least, not yet – but the existing alternative (the ONS’s Labour Force Survey) doesn’t separate out graduates in quite as much detail.

The HESA report takes the easiest possible approach to combining these into a single variable – each question is a five-point Likert scale, so you take an average score and divide by three. The score used to produce these averages come from graduates in full time employment only, so we skip past the issue with “current activity” nicely.

I’ve got a lot of time for radical simplicity, but to do this you need a reasonable sample size. You can get a reliable number using a large group of students, but you need to be very careful when subdividing. So, for instance, if you wanted to compare providers these would need to be reasonably sizable ones.

Let’s do that!

You’ve read this far, so let’s have a go at applying this based on published data about providers. Here are scores presented for reasonably sizable universities and colleges only (FECs and alternatives are filtered out), on a map.

[Full screen]

(Why is it a map? To look at regional impacts, and to avoid it being a league table)

You will note that there’s not a huge level of differentiation between providers – we’re pretty much skewing towards four out of five in each case, the few minor outliers are generally smaller colleges and specialist providers.

What does this tell us? At this level of segmentation, graduates are generally in decent jobs in the narrow sense of this particular bit of a wider definition. I would confidently expect an effect based on industry and on subject of study – and we would also need work on including those in part-time work.

HESA are keen to move forward with this as a published metric based on Graduate Outcomes – and consultation with stakeholders and data users will determine whether this happens and at what level of granularity. Even given the caveats detailed above, I’d argue for publication at provider level to forestall some of the less wellbeing aware metrics that are frequently touted in this area, and at top level SIC to gain access to issues within particular industries that employ graduates.