A cut-out-and-keep summary of the claims most commonly made about the harm that Brexit has supposedly done to UK GDP, their sources, their methodology and – most importantly – their flaws
THERE IS NOW a bewildering range of estimates for the harm that Brexit is supposed to have done to the UK economy, or the benefits of rejoining the EU. Indeed, pro-EU accounts often present several inconsistent figures as if each were established facts – even in the same post!
Here is a brief rundown of the most popular claims, their source, the methodology, and the flaws.
- “4 per cent hit to productivity” / “4 per cent hit to GDP”
- “15 per cent reduction in trade”
- “£100 billion”
- “2.2 per cent of GDP” / “£25 billion”
- “8 per cent of GDP” (based on “doppelgängers”)
- “6 per cent of GDP” (based on micro-level modelling)
1. “4 per cent hit to productivity” / “4 per cent hit to GDP”
Claim: “Brexit will reduce long-run productivity (specifically, GDP per head) by 4 per cent relative to remaining in the EU” (sometimes interpreted as a “4 per cent hit to GDP”).
Source: this was the initial assumption made by the OBR before the UK left.
Methodology: the 4 per cent figure was simply a crude average of the results of 13 external studies.
Flaws: this is not original analysis by the OBR. The 13 external studies were all done before the final shape of the exit agreement was known. They also used a variety of different models and assumptions, most of which now look too pessimistic. Even then, 9 of the 13 studies suggested the impact would be less than 4 per cent.
2. “15 per cent reduction in trade”
Claim: “both exports and imports will be around 15 per cent in the long run than if the UK had remained in the EU” (or a similar fall in the trade intensity of the UK economy).
Source: this was another early assumption made by the OBR.
Methodology: the 15 per cent was based on the average estimate of a number of external studies that looked at the impact of leaving the EU on the volume of UK-EU trade.
Flaws: the OBR’s 15 per cent figure covered total trade in goods and services with the entire world, not just goods trade with the EU. A shock of this magnitude to the UK’s global trade had always looked implausibly large, and it has indeed failed to materialise. In particular, the UK’s trade intensity (or “openness”) has continued to track that of similar countries, such as France. (For more on the UK’s trade performance since Brexit try this piece by David Paton.)
3. “£100 billion”
Claim: “Brexit is costing the UK £100 billion a year in lost output”.
Source: analysis by Bloomberg Economics published back in 2023. (At the time, £100 billion was about 4 per cent of UK nominal GDP, so this often conflated with the OBR’s 4 per cent.)
Methodology: a simple “counterfactual” which assumed that if Britain had stayed in the EU then UK GDP would have tracked other G7 economies in exactly the same way as it did before the vote to leave.
Flaws: for a start, this analysis is now three years out of date. More fundamentally, it assumed that Brexit was the only significant shock impacting any G7 economy in the period since the vote to leave in 2016, ignoring many other factors including Covid and the energy crisis. This is obviously nonsense (also see the discussion of the NBER’s 8 per cent figure in point 5 below).
4. “2.2 per cent of GDP” / “£25 billion”
Claim: “forming a new customs union with the EU would boost the UK economy by 2.2 per cent of GDP” or “the UK-EU reset would boost GDP by 2.2 per cent”. This could equate to around £25 billion in additional tax revenues.
Source: the 2.2 per cent figure was drawn from a February 2025 report by the consultancy Frontier Economics, commissioned by the pro-EU campaign group “Best for Britain”. It has also been widely cited by the Liberal Democrats.
Methodology: the report modelled “deep alignment” in both goods and services and suggested that this would increase UK GDP by 0.3 per cent-0.4 per cent over one year. The 2.2 per cent was a ‘long-run’ estimate which simply extrapolated the one-year figures using some heroic assumptions about the links between trade intensity and productivity.
Flaws: many! The report modelled something that is simply not on the table – regulatory alignment based on ‘mutual recognition’, with the most favourable results assuming that this applies to both goods and services (so not, in fact, the same as a ‘customs union’).
Even then the numbers are dodgy. Specifically, the report assumed that a one percentage point increase in trade openness would boost GDP by as much as 0.5-0.7 per cent. That would be an implausibly large boost for an economy like the UK which is already relatively advanced and open.
5. “8 per cent of GDP” (based on “doppelgängers”)
Claim: “Brexit has already reduced UK GDP by as much as 8 per cent”, with larger impacts on investment in partlcular.
Source: an NBER working paper published in November 2025.
Methodology: the 8 per cent figure was a “top down” estimate derived by comparing growth in UK GDP per capita with the averages of several different groups of other countries. These groups were assumed to be good proxies (or “doppelgängers”) for what would have happened to the British economy if the UK had remained in the EU. (Other variations of this approach have produced numbers of 5.5 per cent, and even more than 10 per cent.)
Flaws: the doppelgänger approach simply cannot separate out the impact of Brexit from other reasons why some economies have grown more quickly (or slowly) than the UK since 2016.
These alternative explanations include the impact of the Covid pandemic and the energy crisis, and many others. For example, the US has benefited not just from relatively low energy but also from a large fiscal stimulus and the AI boom.
Care should be taken too when comparing the UK’s economic performance since 2016 to that of countries hit hard by the euro area debt crisis of the early 2010s. The numbers for Italy, Spain and Greece in particular have been flattered by catch-up growth. The 8 per cent figure is implausibly large if you compare the actual performance of the UK economy to better matches such as France or Germany.
(For more on the NBER analysis see “Three reasons why you should never trust a doppelgänger”.)
6. “6 per cent of GDP” (based on micro-level modelling)
Claim: “Brexit has already reduced UK GDP by 6 per cent”
Source: this figure also comes from the NBER working paper published in November 2025.
Methodology: the 6 per cent figure is based on a “bottom up “ approach using firm-level data from the Bank of England’s Decision Maker Panel (DMP) survey. But just like the “top down” approach used to arrive at the 8 per cent figure, any divergence in the performance of firms based on their pre-referendum exposure to the EU was assumed to be solely due to Brexit.
Flaws: several. The DMP is a relatively large survey, but the respondents are not necessarily representative of the UK economy as a whole. More importantly, any divergences in the performance of individual firms depending on their exposure to the EU may also be picking up the relative weakness of major European economies over this period (especially Germany), or other factors such as the impact of the UK’s relatively high energy costs on the competitiveness of goods which are mainly exported to the EU.
Finally, while the prospect of an increase in trade frictions with the EU will clearly have had some negative effects, these should prove to be (mostly) temporary as Brexit uncertainty fades and firms adjust to the new trading arrangements.
Indeed, the DMP itself suggests that Brexit has dropped well down the list of concerns for most firms. Similarly, the latest Deloitte survey suggests that CFOs are now more worried about “economic weakness in the euro area and the possibility of a renewed euro crisis” than they are about the “effects of Brexit or a deterioration in UK-EU relations”.
Established in 2006 and celebrating twenty years, ThinkScotland is not for profit relying on donations to publish our wide range of opinions. You can subscribe to our FREE newsletter here – share and follow us on ‘X’ here – like and comment on facebook here and make a donation here. Every contribution helps to meet our overheads.
This article was first published on Julian Jessop’s Substack which you can subscribe to free here.
AI-Generated Image of a knockout punch by Uolir from Adobe Stock
