The cost of Brexit: March 2021

We estimate that leaving the single market and customs union had reduced UK trade by 11 per cent in March 2021. That is on top of a 10 per cent hit to trade between the referendum and leaving the single market.

Last month, our cost of Brexit model showed that leaving the single market and customs union had reduced the UK’s total goods trade by 5 per cent in February. Using the data for March, which was released today, we estimate that goods trade is now 11 per cent, or £7.7 billion, lower. There are two reasons why our estimate worsened between February and March. Trade growth in the countries that make up our ‘doppelgänger’ UK outstripped Britain’s, widening the gap with our modelled economy that stayed within the single market and customs union. The ONS has also revised February’s trade data downwards. It is important to remember that monthly trade data is volatile, so it will take several more months to be certain about the effect of Brexit on the level of UK goods trade, but it is becoming clearer that the impact cannot be dismissed as temporary.

To estimate the effect of single market and customs union exit, we use trade data from other advanced economies. An algorithm chooses – from a ‘donor pool’ of 22 advanced economies – a smaller selection of countries with economic characteristics that most closely matched those of the UK over the last decade. Those countries are combined into a ‘doppelgänger’ UK, with the relative weighting of the selected countries chosen to create the smallest possible deviation from the real UK goods trade data between 2016 and 2019. By comparing the UK’s actual goods trade performance from January 2021 to that of the doppelgänger, we can assess how leaving the single market and customs union has affected Britain’s trade in goods. (For more details on how the model works, see the estimate for January.)

COVID-19 does not significantly affect our model, because we only use it to evaluate the UK’s performance from January 2021, when goods trade in advanced economies had largely recovered to pre-pandemic levels. The countries in our doppelgänger UK are chosen using pre-pandemic data, which also reduces the impact of the virus on our estimate. 

Our second estimate compares the UK data to a doppelgänger that did not vote to leave the EU. Had the UK voted to remain in the EU in June 2016, its trade would be significantly higher. The depreciation of sterling after the referendum hurt UK trade, instead of helping British exporters as many speculated. The unwinding of UK-EU supply chains in anticipation of trade barriers also reduced trade flows. Our second doppelgänger is made up of the countries whose goods trade had the smallest possible difference from the real UK data between January 2012 and the referendum in June 2016. According to this estimate, the Brexit process had reduced UK goods trade by £8 billion, or 10 per cent, between the referendum and the end of 2020.

John Springford is deputy director of the Centre for European Reform.