Trade in services under regulatory barriers: Evidence from UK banking

Non Technical Summary

A growing body of academic research highlights the significant impact of Brexit on UK trade. Yet, trade in financial services in general, and banking services in particular, remain underexplored. One reason is the limited availability of detailed data despite its importance for the UK economy. This paper provides new evidence on the effect of Brexit on trade in banking services.

UK banks that lost passporting or were more exposed to the EU saw a significant fall in their activity with the European Economic Area (EEA) non-financial sector

We analyse the impact at the bank-level using confidential statistical data from the Bank of England. Our analysis spans the period 2014 to mid-2024.

We focus on the loss of passporting following the new trade arrangement that came into effect in early 2021. Passporting allows financial firms authorised in one EEA country to provide services in other EEA countries, either cross-border or through branches, with minimal additional authorisation. We define sets of banks established in the UK that were affected or unaffected by the loss of passporting to measure their relative impact on the stocks of loans provided and deposits taken. When the UK was a member of the EU, both banks incorporated in the UK and branches of EEA banks could operate cross-border through passporting. By contrast, UK branches of non-EEA banks never had passporting rights.

We find that the banks that lost passporting authorisation experienced a 40-50% additional fall in stocks of loans to, and deposits from, the non-financial sector in EEA countries, relative to their value in 2016Q1 and relative to banks without passporting authorisation and activities with non-EEA countries. While the additional fall is unsurprising, the magnitude of the relative effect suggests that regulatory barriers after the loss of passporting had a significant impact on cross-border service provision.

In an alternate specification, we look at the effect based on the banks’ exposure to the EEA before the referendum (i.e. banks with a large share of their non-resident activity with the EEA pre-referendum). Consistent with the passporting results, we find that the more the bank was exposed to the EEA pre-referendum, the larger the decline in activity with the EEA, and this effect starts after the referendum itself.

This fall is also significant for activity with the financial sector in the EEA

We study the impact on UK banks' lending to and deposit-taking from other banks and financial corporations in partner countries. UK banks that lost passporting saw a significant decline in deposits from banks in the EEA and a significant decline in both lending to and deposit-taking from financial corporations in the EEA relative to UK-banks that did not have passporting authorisation and relative to activity with non-EEA countries. Overall, our results suggest that the UK's banking activity has fallen since the Brexit vote in 2016.

Multinational banks operating in the UK could not reduce the impact of new trade barriers by expanding the activity of affiliates in the EEA

Multinational banks with operations in the UK may adjust to higher UK-EU trade barriers by shifting banking activity to their affiliates within the EEA. To assess whether this happened following Brexit, we estimate the impact on lending and deposit-taking of UK banks with banks in the same company-group (henceforth intragroup) located in other countries. We find that banks that lost passporting authorisation reduced their lending to intragroup banks in the EEA significantly, relative to those that did not have authorisation and activities with non-EEA banks, suggesting that intragroup activity was not used to transfer capital to restricted markets and that cross-border barriers mattered in such transactions as well.

To further investigate if there was instead an expansion of banking affiliates of a given company to access the market with increased barriers, we use Historical Orbis to obtain information on all intragroup banks under the same ultimate owners as the UK banks, located in other countries, over the period of our analysis. We find that there is an increase in the number of foreign affiliates in the EEA of banking groups affected by the loss of passporting in the UK relative to banking groups that were unaffected. However, these intragroup affiliates located in the EEA did not have any relative increase in their lending or deposit-taking activities.

Overall, our study provides an example of what complex regulatory barriers to services look like and suggests that the increase in regulatory barriers after Brexit materially reduced banking integration between the UK and the EEA.

This paper is also published by the Bank of England, Staff Working Paper No. 1,181.