I often come across ‘zombie statistics’ for my collection of ‘bad Brexit takes’. Here’s a classic example for those not familiar with the genre.

This week the Guardian published an article by Polly Toynbee which included the claim that “the Office for National Statistics says Brexit cost the UK £1m an hour in 2022”. This was then reposted as fact by many of the usual suspects, notably Liz Webster, Alastair Morgan, and Richard Corbett. But only a cursory check is needed to confirm that it is garbage.

For a start, this figure does not come from the ONS. That is simply false.

Instead, it is based on ‘research’ by ‘ParcelHero‘ (I’ve looked them up and they are an international courier company), which was first reported by CityAM back in 2022.

So how did they get £1m an hour?

First, they dug out an estimate of £11bn for the UK’s net contribution to the EU budget in 2018 (sic). Not paying this is then credited as the only benefit of Brexit.

Then they subtracted the fall in the value of UK goods exports to the EU in 2021 (sic) compared to 2018 (sic), which they said was £20bn. This was then attributed entirely to Brexit (Covid, anybody?), and debited as the only cost.

Finally, these apples and oranges were combined to arrive at a net figure for the annual cost of Brexit of £9bn, which would indeed work out at about £1m an hour.

It should be clear by now that this approach is just nonsense economics, and that no serious commentator would use the results.

Moreover, there have been several attempts to kill off this statistic before.

BBC Verify had a go back in 2023, when Alastair Campbell claimed that “the ONS have also said that Brexit is costing us a million pounds a day (sic)” in a Radio 4 debate with Jacob Rees-Mogg.

As the BBC noted, “this is not a good way to measure the impact of Brexit on the economy because it allocates the entire fall in UK exports in 2021 compared with 2018 to Brexit, when there were other things going on, notably the Covid pandemic; these figures are out of date – we already have trade figures for the first nine months of 2022, which show considerable increases; and it is only looking at exports and not, for example, imports and investment”

Apparently, Campbell had been fed this statistic by the European Movement, which has also used it before (though even then he couldn’t get the line right!).

Unfortunately, there is a ready market for this kind of crap, as well as plenty of deeply unserious people willing to push it. And this isn’t even the most egregious example of a ‘zombie statistic’ deployed by Project Rejoin.

An even worse one was Mark Carney’s claim in 2022 that “in 2016 the British economy was 90% the size of Germany’s. Now it is less than 70%” (which I debunked here).

Or the New Statesman’s claim in 2023 that the Irish economy is ‘booming at the expense of Brexit Britain’, on the basis of Ireland’s notoriously flawed GDP data (debunked here).  

Does this matter? Many people have tried to quantify the economic impact of Brexit in different ways, and some have come up with bigger numbers using methods that are at least a little more credible. I am happy to debate, for example, the OBR’s assumption of a 4% long-term hit to productivity (reviewed here).

But everybody should be fed up with the lazy use and re-use of dodgy statistics, whether they supported Leave or Remain. The fresh tsunami of fake news is certainly doing nothing to improve the quality of the current debate on future relations between the UK and the EU.

• Applied economics
• Brexit
• Politics
• Trade policy
• UK economy