Where London’s Banking Jobs Went After Brexit

Nearly ten years after the referendum, there are two sets of numbers on London, and people tend to quote one while ignoring the other.

The first says London has held its position. New Financial’s international financial centres index (July 2026) still puts the UK at more than four times the size of the next largest centre in Europe, and alongside the US as one of only two global centres when measured by actual market activity. UK international financial activity grew 20% in real terms between 2015 and 2025, a little ahead of the 17% global average.

The second says a great deal left. New Financial’s 2021 count found more than 440 firms had moved business, staff, assets or legal entities to the EU, taking around £900bn of bank assets, about a tenth of the UK banking system, along with 7,400 identified roles. The ECB now supervises 14 subsidiaries of non-EU banks holding €1.9 trillion, roughly 7% of everything it supervises (ECB, March 2026).

Both are accurate. What reconciles them is that most of what moved was legal structure and booking capability rather than the market itself, and a lot of siting decisions go wrong by treating the two as the same thing. It led to an increase in VP-level hiring as these bankers count as 'officer-level' employees and every deal has to have one meaning that VP-hiring amongst integrated banks increased to cater for this specific need for experienced talent.

Where the firms went

Frankfurt and Amsterdam get most of the attention, but they were not the main destinations. Dublin took 135 firms, about a quarter of the total, with Paris on 102 and Luxembourg on 93. Frankfurt took 62 and Amsterdam 48, and nearly 70 firms opened in more than one EU city.

What each city received mattered as much as the count. Around 60% of the firms arriving in Frankfurt were banks, while almost two thirds of those choosing Amsterdam were trading venues, exchanges and brokers. Frankfurt gained regulated banking entities; Amsterdam gained market infrastructure, which is important to how European markets function but employs relatively few people.

For anyone deciding where to put a team, the more useful question is which city holds the function you need, not which one “won”. A tally of firms says little on its own, since a Luxembourg fund management company and a Frankfurt bank subsidiary count the same.

What the regulator required

The ECB pushed hard on how incoming banks were structured and published what it found. Its desk mapping review covered 264 trading desks across seven banking groups and required supervisory action on 56 of them, or 21%. Around 70% of the desks assessed were running back-to-back booking models, with roughly another 20% split between entities (ECB Banking Supervision, May 2022).

The supervisor wanted local risk ownership: desk heads, risk takers and control functions sitting inside the euro area entity, with the infrastructure to support them. Coverage bankers, research and sales serving the same clients were largely left where they were. That is the difference between a build of a few dozen people and one of several hundred, and it is worth asking of any competitor’s Frankfurt announcement how many of the roles carry a risk mandate.

Why London kept the rest

Some businesses are expensive to move because of what they are attached to. Around two-thirds of euro-denominated derivatives trading still takes place in the UK. About a fifth of EU domiciled investment funds are managed from the UK, and in the three years to 2024 a quarter of EU venture capital investment involved a UK investor, while 28% of VC investment into UK companies involved an EU one (New Financial, December 2025).

The two markets have stayed closely linked in practice even as the legal arrangements have grown more complicated, which goes some way to explaining why the relocation figures levelled off. Once a bank had satisfied its supervisor, there was little reason to send more people.

The constraint on the continent

The bigger question for London may be less about whether Frankfurt or Paris takes its business than about whether the EU develops capital markets deep enough to generate more of its own.

The ECB’s own reading is mixed. Financial integration has improved since late 2022, but equity market integration has slipped back, cross-border equity investment has stalled, and foreign direct investment within the euro area is at historically low levels, with households still holding much of their savings in deposits (ECB, May 2026). Banking also remains more fragmented than in the US: the eight largest US banks hold 55.2% of banking assets, against 41.3% for the eight largest in the euro area.

Headcount points the same way. EU banking employment fell 0.8% in 2025, with 16 of 27 member states reducing staff, and the number of bank offices dropped 2.6% (ECB, June 2026). Much of that is likely retail restructuring rather than wholesale changes in team composition, but it is not the picture of a market absorbing large numbers of London roles.

What could change

None of this means the current balance is fixed. From January 2027, banks from outside the EU will need a physical EU presence for deposit taking and lending, and the ECB expects that to raise the non-EU bank assets it supervises materially. That is a second wave of structural relocation with a fixed start date, and it may move more real activity than the first.

UK domestic markets also grew just 3% in real terms over the same decade. A centre that relies so heavily on international business is more exposed to decisions taken elsewhere than its headline ranking suggests.

What this means for siting a team

The evidence suggests London lost legal entities, booking capability and a defined group of risk takers, mostly to Dublin, Paris and Luxembourg, and that the first wave has largely run its course. Planning on the assumption that activity will keep drifting east risks building for a shift that hasn’t happened.

For most European product lines, a sensible working rule is to place risk where the supervisor requires it and everything else where the talent is. In practice that still usually means London for scale and Paris for corporate and sponsor coverage of continental clients. French sponsors, just like other EU nations's capital holders, don't tend to want to be covered from London by Brits (generally).

The harder part is knowing whether a euro area team can be staffed at all. That depends on how many people with the right mix of language, product and regulatory experience live within reach of a given city, and the answer is often smaller than expected. An applicant list shows who was interested in a Frankfurt role; it doesn’t show whether the role can be filled. We’d be curious to hear how others are thinking about the 2027 deadline and whether it is changing where they plan to build.

‍