In May 2025, HM Government released a white paper called “Restoring Control over the Immigration System”. One policy proposed in this paper was a levy on international student fees, to be “reinvested into the higher education and skills system.” Although no details were given, the technical annex proposes a 6% flat levy on all international fee income.

The technical annex went on to project the impact that this price rise would have on international student numbers. To estimate this, the Home Office uses a London Economics estimate of price elasticity of demand of EU students from 2021. However, as the technical annex makes clear, this assumes that non-EU students in 2025 have the same price sensitivity as EU students in 2021. In order to give an accurate assessment of the impact this proposed levy could have on international student numbers, we need updated and extended data. 

This work extends and updates previous modelling on elasticity of demand from international students attending UK universities (London Economics, 2021). It provides updated assessments as of 2025, and in particular focuses on non-EU student demand, representing 90% of all current international students in the UK.

We find that:

• Price elasticity of demand among non-EU students is greater than that found for EU students, at -0.57 in the short run
• When we break this down into undergraduate and postgraduate demand elasticity, this is almost three times the size of the estimates of elasticity that are currently being used to predict the impact of price rises on demand. That means, if government models on the basis that 2021 EU student price elasticity is the same for non-EU students in 2025, they are drastically underestimating the impact of price increases on demand; almost three times the number of non-EU students would be put off coming to the UK than predicted by the 2021 estimate, in first year of any price rise. 
• In the long-run, we do not find a statistically significant effect of demand elasticity from non-EU students at an aggregate level. Even so, it would take 5 years, all else being constant, for the sector to recover half of those lost student numbers. However, when decomposing elasticity into that of undergraduates and postgraduates, we do find statistically significant results in both the short term and long term, for undergraduates (-0.54 and -0.25) and postgraduates (-0.81 and -0.46) respectively.
• We do not find statistically significant results in replicating London Economics methodology for EU students in the run up to 2021; we hypothesise this because of relatively lack of data and volatility in recent years. We utilise London Economics data for EU demand elasticity when making wider calculations of impact on the sector.
• We project that if any new levy means fees increase by 6.38% in real terms, around 4.68% fewer non-EU students will enrol, the first year that any levy is introduced. This equates to a decrease of 5.17% of non-EU postgraduate students and a decrease of 3.38% of non-EU undergraduates. We also forecast that both undergraduate and postgraduate non-EU enrolment numbers will permanently shrink in the long run, by 1.53% and 2.94%, respectively. 
• Combining the non-EU and EU estimates of decreases in demand, we project that, with a 6.38% rise in international student fees, the sector will lose over 16,100 international students in the first year the levy is introduced.
• In financial terms, such a decline in student numbers would see an approximately £240 million loss in fee income in the first year the levy was introduced. Over 5 years, the total number of international student enrollments could decrease by over 77,000 students, and could cost the sector around £2.2bn.
• Were the reduction in international student revenue paid for entirely by reducing subsidised domestic students, there would be 33,000 fewer domestic places in Year 1 of any levy. Over 5 years, this could be around 135,000 fewer domestic places available. 
• Alternatively, if the reduction in revenue were paid for from the research budget, the amount of money available for research and innovation could shrink by 1.53%. To give a sense of scale, over five years this is roughly equivalent to 2.9x the budget for the new Edinburgh supercomputer, or the entire increase to the defence budget announced in the 2025 Spring Statement.
• Regional impact of such an economic contraction is unsurprisingly focussed around university heavy constituencies. The ten most affected constituencies would lose almost £40m each annually in GVA contraction; these constituencies are spread across the country, and cover London, Coventry, Manchester, Glasgow and other cities where student populations are high.

This work has been authored by Rhiannon McQuone and Jonathan Simons, with economic analysis and support from Scott Corfe, Ben Savours, and Dr Vikram Bahure.

You can read the report here: ISL Modelling – Final Report