How education related exports and transnational education activity data is made

Welcome to the inside of the sausage machine

David Kernohan is Deputy Editor of Wonkhe

Unbelievable as it may seen, I do get stick occasionally for banging on about data definitions and methodology rather talking about what the data actually says.

In all honesty, some data (including this publication) is just so arbitrary that the actual numbers don’t really matter all that much. But in the interests of keeping the peace here’s a paragraph from my esteemed colleague Michael Salmon:

“The year saw revenue from education-related exports and transnational education (TNE) reach £27.9bn, up from £26.2bn in 2020 – the current UK International Education Strategy aims for this figure to reach £35bn by 2030. Education exports include international students’ tuition fees and living expenses in the UK – these are estimated at £20.65bn for 2021. Pathway providers are assessed to have contributed £0.45bn, and research grants and contracts to have brought in £1.46bn, both figures a slight decrease on the previous two years. The share of TNE income attributable to higher education is £0.89bn, a rise from £0.77bn in 2020.

Sounds guardedly positive and even encouraging in difficult geopolitical and financial times, doesn’t it? Let’s head for the methodology.

Anyway – the first thing you will spot is that 2021 was a long time ago. The original aim was to do 2021 and 2022 on the same day, but because of – you guessed it – problems with 2022-23 HESA Student data this has not been possible. When you look back at the data think back to 2021 – we were just coming out of the main period of Covid restrictions, with every chance of more to come, and international travel was subject to restrictions and otherwise unpopular.

It’s also notable that all these figures are based on exports only – there is no adjustment at all for costs incurred in delivering a service overseas.

There’s been a few tweaks to the fee income calculation (some scholarships are now included, EU-based student fees for those starting after 2021 are now included – with the old cost to government, or RAB, estimates removed). We’ve also updated living cost calculations based on the latest (2021-22) Student Income and Expenditure Survey – but there’s some fairly arbitrary assumptions about the number of weeks a year students of various nationalities and backgrounds are resident in the UK)

On intellectual property income, it is estimated that this constitutes 37.5 per cent of all intellectual property related income as reported in the HESA HE-BCI survey – this figure comes from a London Economics report published in 2011 based on 2008 data! The same paper provides figures for other overseas income – which has simply been uprated based on the changes in performance between 2004-05 and 2008-09, and even back then it was based on 14 survey responses. And as DfE notes “It is… not clear what is covered in this category”.

Finally, for higher education, pathway provider income (programmes that help to prepare overseas students for study at a UK university) is estimated based on a survey of six large providers (CEG, INTO, Kaplan, Navitas, Oxford International, Study Group) conducted by one of the participants (Kaplan).

Incredible as it may seem other bits of the data are even more arbitrary (we don’t know anything about the number of EU domiciled students at FE colleges, there’s no estimate at all for living costs for international students at independent schools, English language study data comes from an actual magazine, education equipment is based on 2020 data (the latest available), and another London Economics estimate from the 00s drives the broadcasting calculation). And do add to this the fact that this is an experimental methodology that is not directly comparable to previous years of data.

None of the above, of course, will stop this data being widely cited in parliament and beyond. Data driven policy, eh?

Leave a Reply