Much of the conversation about regulation in England over the past three years has been, albeit obliquely, about subjects of study.
From Gavin Williamson’s grisly dispatch-box delight at defunding “meeja studies”, to 90s pop icon Jarvis Cocker weighing in on the value of “the arts”, to various campaigns around subjects as diverse as archeology, languages, and social geography – it feels like everything outside of the STEM quintilateral has been facing some form of existential crisis.
Augar – of course – had thoughts on the matter. Recommendation 3.5 suggested:
Government should adjust the teaching grant attached to each subject to reflect more accurately the subject’s reasonable costs and its social and economic value to students and taxpayers.
The background to this was the observation that the English funding model allows the government to provide a selective subsidy to subjects where average earnings are such that the full value of the loan (plus interest) would not be repaid by graduates. It’s an argument repeated in the government response, but the proposed solution is couched in the more modern (well, the 2019 manifesto rather than the 2018 review launch) language of “low-quality provision”.
Close reading
As this language has overtaken the discourse about subjects (or even courses), the sector has tended to read subject into such statements. In reality, the OfS proposals on B3 would penalise a physics course with a high dropout rate in the same way as it would clamp down on a printmaking course where graduates faced low salaries and unskilled labour. Though subject of study correlates with earnings and destination (as a million IFS studies aptly demonstrate) the recent government approach has been to downplay these statistical – if not necessarily causal – links.
Why? Well, I could be unkind and suggest that the government is unable to define a subject. It would follow the pattern through from the initial framing of a low quality “course”, where ministers clearly did not realise how little information existed at course level or what a course actually was. Though your subject of study feels to the layperson like a neat way of splitting provision at a university into meaningful chunks, Wonkhe readers will know that this is not the case.
Subjects within subjects
Even in the much discussed PROCEED metric we never get beyond CAH level 2 – “computing” ranges from AI to multimedia to software design. In the most recent LEO release we get a lot more subject fidelity – male artificial intelligence graduates from Liverpool John Moores earned a median salary of £24,800 five years in, whereas male software designers were looking at £27,600. Salary has fallen from favour as a metric almost entirely, but it’s not controversial to suggest that similar differences lurk beneath the top level CAH splits in graduate outcomes.
So proposed government intervention at subject level extends largely to chucking more money at STEM (the policy of every government since Atlee) and the mess around student number controls – which is far more focused on cutting government spending than squeezing undesirable subjects of study off campuses. What is likely to result is something akin to 2011’s core and margin system – with broad subject area caps linked to historic recruitment patterns and demographics that can be selectively released to allow for growth.
Three Skinners
This selectivity could be controlled by metrics or ministerial whim – quantifiable, societal, or strategically important subjects in other words. The sparse implementation system proposes the curious anomaly that imposing these controls may not be the role of the OfS – coherence to whose B3 standards is chucked in very much as an afterthought.
It’s been a long way round, but redesigning the pre-2011 state funding regime to take account of fees appears to be where we have ended up. The experiment in using the invisible hand of the market rather than the dead hand of the state to optimise subject delivery has clearly failed – it is, as (Seymour) Skinner put it, the children who were wrong, or as (B.F) Skinner would argue operant behaviours triumphed over respondent behaviours. (If you were wondering, studying English at what is now Birmingham City University was an “absolute life-changer” for Frank Skinner)
On this basis, we disappoint Williamson and charm Cocker, which is largely as it should be. The arts and media studies are safe for now. Perhaps.
