WEEKEND READING: The £265 billion illusion: part one

Join the Policy Institute at King’s College London and HEPI on Tuesday 15 September, from 6.30pm to 7.45pm, at Bush House, London, for a free event exploring the future of higher education in England. Drawing on Professor Sir Chris Husbands’ report, New choices: Revisiting futures for higher education in England, the discussion will consider how financial pressures, generative AI, changing student expectations and declining public trust are reshaping the sector – and the choices universities and government need to make now to secure a sustainable future. Register here.

This blog was kindly authored by Paul Marshall, a former director of Pragmatix Advisory.

Every year, universities across the UK publish multi‑billion‑pound estimates of their economic impact. A 2024 report by London Economics for Universities UK put the combined economic impact of UK higher education at £265 billion. The figures are impressive and the conclusions familiar: this institution is a major asset to its city and region. Yet these numbers rarely resonate beyond the institution itself nor are they routinely taken into account by policymakers. Communities around major campuses remain deprived, politicians remain sceptical, and the public, if it notices at all, moves on quickly.

The reason is not that universities aren’t economically significant: they are, and economic impact assessments serve a legitimate purpose in demonstrating that. The problem is that the standard methodology is designed to produce an impressive headline figure, not to answer the questions that actually matter: who benefits, by how much, and from where? This is not because the figures are necessarily wrong. Rather, it is because they measure scale rather than distribution. They tell us how much economic activity is associated with a university, but very little about who benefits from it, how those benefits are distributed, or whether they contribute to wider prosperity within the places universities claim to serve. The issue is not whether to measure economic impact, but whether the current approach is measuring the right things.

The model and its assumptions

The dominant approach uses input‑output modelling: add up what the university spends and earns, apply multipliers to capture knock‑on effects through the supply chain and employee spending, and arrive at a total GVA and employment figure. Despite the proprietary branding consultancies apply to their models, the inputs and coefficients are broadly standardised as they all draw on the same publicly available ONS multipliers. What varies is not the science, but the assumptions.

Four frequent assumptions deserve particular scrutiny.

First, employment is treated as homogeneous. A professor and a catering worker on a zero‑hours contract both count as ‘one job’. The model captures neither wage distribution nor the difference in local economic impact between high‑paid and precarious work.

Second, the geographic boundary of ‘local impact’ is typically drawn to maximise the figure. Supply chain and employee spending effects are attributed to a wide geographical area without testing whether local suppliers actually capture the spending.

Third, national multipliers are used in place of localised ones. National averages are skewed toward London and the South East, where economic output is highest. For post‑industrial cities, which host many universities with the strongest civic missions, the appropriate local multiplier would be materially lower. Impact figures in England’s old industrial heartlands are therefore almost certainly overstated, and those in London possibly understated.

Fourth, and most significantly, these models make no adjustment for factors that HM Treasury’s Green Book requires in any credible economic evaluation: additionality (would the activity have happened anyway?), deadweight (what would have occurred in the university’s absence?), leakage (how much spending leaves the local economy?), displacement (does university activity crowd out other actors?) and substitution (are university jobs replacing jobs elsewhere?). Assessments cited in policy lobbying as rigorous economic evidence would, in most cases, fail these Green Book tests outright.

A number that obscures more than it reveals

What these choices produce is a single aggregate figure that tells us little about the real economic relationship between a university and its place: who captures the gains, who bears the costs, or whether the institution is performing well or merely operating at scale. Indeed, the methodology often conflates institutional scale with institutional performance. A large university will almost always report a larger economic impact than a small one, regardless of whether it generates stronger local outcomes.

Consider two institutions with very different missions, such as UCL and London South Bank University. UCL’s calculated economic impact is likely to be substantially larger, reflecting its scale, research intensity, and international reach. But that tells us very little about whether it generates greater civic value within its immediate locality. LSBU’s contribution to social mobility, graduate retention, and civic engagement in South London may be far more significant than the GVA gap implies, yet the standard model has no way of capturing this.

There is also a distributional dimension the model cannot see. An institution can be simultaneously the largest employer in a city and a net contributor to its inequality if the local jobs it creates are low‑paid, the graduates it produces leave, and its procurement spending flows to national contractors rather than local businesses. The evidence for this is considered in the next section.

The question the model doesn’t ask

Economic impact assessments are not going away, and there are legitimate reasons for that. Universities must make the case for public funding and compete for staff, students, and partnerships. But these pressures have produced a structural tendency toward optimistic framing at the expense of rigour.

The fundamental problem is not that the standard model is wrong. It is that it is answering the wrong question. “How large is the economic activity generated by this institution?” is legitimate, but it is not the same as “Does this institution make its place more prosperous, and for whom?”

In an era of place-based industrial strategy, devolution and regional rebalancing, with ‘good growth’ (broadly, growth that is both productive and inclusive) now embedded in the policy lexicon, the second question is the one that matters. Universities that can answer it with disaggregated evidence on graduate retention, local procurement spend, wage distribution, and community outcomes will be better placed to demonstrate their public value than those that simply produce a larger GVA number.

Good for growth – but whose growth?

Universities are undoubtedly forces for good when it comes to economic growth. But the evidence reviewed here suggests that this growth is most reliably felt at a national level or in areas that are already economically advantaged. In deprived communities, universities too often coexist with – and in some cases exacerbate – the disadvantage around them. The HESA access data tells one story; the ward-level deprivation indices for the same postcodes tell another.

Middlesbrough illustrates the point sharply. A 2022 report by independent economists New Skills Consulting found that Teesside University contributes over £247 million in GVA to the economy each year, supporting almost 3,300 jobs, with £161 million of that GVA attributed specifically to the Tees Valley. Teesside’s campus sits in Middlesbrough town centre – physically embedded in the community it claims to benefit. Yet Middlesbrough is the second most deprived local authority district in England on the 2025 Index of Multiple Deprivation, a position of entrenched disadvantage it has held since at least 2010.  That contrast is not an accident of geography. It is the predictable outcome of a model that measures scale rather than distribution, and aggregate activity rather than local benefit.

None of this should be read as an argument against universities. On the contrary, universities remain among the most important institutions in many local economies, and their contributions to research, teaching, and civic life are real and substantial. The issue is not whether they generate growth, but how that growth is experienced, distributed, and retained locally – and whether the evidence base we use to assess it is up to that task.

If universities are to play a leading role in place-based growth and economic renewal, they need better ways of demonstrating impact than increasingly large GVA figures. The key question is no longer how much economic activity a university generates, but who benefits from it, where those benefits are retained, and whether they contribute to more inclusive prosperity. Until we can answer those questions with confidence, the billion-pound impact headline will remain an impressive statistic – but an incomplete measure of public value.

In tomorrow’s blog, I will provide a tool and mechanism by which universities can provide a more holistic assessment of economic impact, locally, nationally and internationally.