The UK did not switch from factories to finance at a single Brexit moment. The long-run change began much earlier, and the strongest post-2000 evidence shows a broader shift toward professional and digital exports while manufacturing retained important domestic production, supply-chain and regional functions. [Office for National Statistics · 2418] [Office for National Statistics · 2419] [Office for National Statistics · 2420] [Office for National Statistics · 2423]
Key findings
The clearest external-balance result is the scale of the services surplus. In 2025 it reached £203.8bn, only modestly smaller than the £225.6bn goods deficit. Services therefore absorbed most of the visible merchandise-trade imbalance, even though that comparison says nothing by itself about replacement of manufacturing jobs or value added. [Office for National Statistics · 2423]
The second finding is heterogeneity inside services. Real other-business-services exports increased far more consistently than financial-service exports, and telecom, computer and information exports expanded strongly in nominal terms. A “finance replaced factories” narrative misses much of the actual export transformation. [Office for National Statistics · 2419] [Office for National Statistics · 2420] [Office for Budget Responsibility · 2454]
Manufacturing became smaller in relative terms, not a straight-line collapse
ONS industrial accounts put manufacturing at 8.6% of UK GVA in 2024. That share is low compared with earlier decades, but revisions to 2024 manufacturing growth – from 0.3% to 1.5% in the Blue Book process – are a reminder that annual output can recover even while the sector loses relative weight. [Office for National Statistics · 2418] [Office for National Statistics · 2476]
The most defensible interpretation is structural rebalancing rather than uninterrupted industrial decline. The 2008 financial crisis, the pandemic and the 2022 energy shock created very different production episodes, so a simple 2000-versus-2025 endpoint hides the path between them. [Office for National Statistics · 2417] [Office for National Statistics · 2429] [Office for National Statistics · 2442]
Business and digital exports became the long-run growth engine
Telecommunications, computer and information service exports rose from £5.459bn in 2000 to £49.861bn in 2025 at current prices. Real exports of other business services increased from £54.020bn to £183.340bn over the same endpoints. The price bases differ, so the two growth rates should not be compared mechanically; their common message is sustained expansion. [Office for National Statistics · 2419] [Office for National Statistics · 2420]
Other business services matter because they cover a wide set of professional functions – legal, accounting, consulting, research and related activities – that are not dependent on the balance sheet of the City. Real exports of these services to the EU also increased from £51.638bn in 2019 to £66.853bn in 2025. [Office for National Statistics · 2477] [European Commission · 2449]
Services buffered the trade deficit, but that is not a manufacturing replacement rate
Putting the 2025 services surplus of £203.8bn beside the £225.6bn goods deficit gives a useful external-balance comparison. The surplus was roughly nine-tenths of the deficit, which explains why services are central to the UK trade story. [Office for National Statistics · 2423]
The same ratio becomes misleading when moved outside the balance of payments. Manufacturing GVA measures domestic value creation; the goods balance includes imports as well as exports; employment counts people or jobs; FDI measures ownership relationships. There is no coherent economic meaning in adding those quantities into a single “offset percentage.” [Office for National Statistics · 2442] [Office for National Statistics · 2424] [Office for National Statistics · 2425]
Brexit imposed its clearest service cost on regulated finance
The Trade and Cooperation Agreement preserved a framework for services, but it did not recreate financial passporting. Firms serving EU clients had to pay more attention to national authorisations and entity structures, and both UK and EU regulators explicitly prepared markets for the end of automatic single-market access. [Office for Budget Responsibility · 2454] [Bank of England · 2457] [Financial Conduct Authority · 2436] [European Banking Authority · 2437] [UK Government · 2433]
A Bank of England micro study gives the regulatory change a measurable footprint. Banks more exposed to the new authorisation barriers experienced a relative decline of about 45% more in EEA lending and deposit balances. That effect is much more specific than the claim that “all services suffered from Brexit.” [Bank of England · 2457]
Finance remains large, but it no longer explains the whole services story
Finance still generates a major share of the UK services surplus, yet its recent record is mixed. The Treasury has noted that the sector as a whole has not grown in real terms since 2010, and the OBR assessed financial-service exports as 5.9% lower than their 2019 level. [HM Treasury · 2455] [Office for Budget Responsibility · 2454]
That is very different from a collapse of London as a financial centre. UK banks retain huge cross-border balance sheets and the Bank of England continues to judge the system through capital and liquidity stress tests. The evidence supports a story of market-access friction and weaker growth, not disappearance. [Bank of England · 2430] [Bank of England · 2486] [Bank of England · 2487]
Global financial exposure creates income opportunities and shock channels
UK-owned monetary financial institutions held $5.1817tn of ultimate-risk foreign claims in 2025 Q4, while UK-resident MFIs reported $6.4334tn of external claims. The two statistics use different scopes and must not be added together, but each confirms the international scale of UK financial intermediation. [Bank of England · 2486] [Bank of England · 2487] [Bank of England · 2464]
Large foreign assets can support fees, financing and global client relationships. They also make exchange-rate, credit and liquidity shocks abroad more relevant for the domestic system. FDI adds a separate ownership channel, reinforcing the point that a service-led economy can be globally diversified and globally exposed at the same time. [Office for National Statistics · 2424] [Bank of England · 2430]
Export growth does not automatically become jobs, wages or household consumption
Workforce-jobs data are needed to determine where employment actually expanded. A high export value can come from a capital- or skill-intensive activity with limited headcount, while a lower-value domestic service can employ many more people. Export performance therefore cannot stand in for labour-market performance. [Office for National Statistics · 2425] [Office for National Statistics · 2493]
Household consumption is even further removed from export receipts. Real income, inflation, interest rates, taxes and asset prices mediate the transmission. The 2025 consumer-trends data and earnings releases show why a service-export boom can coexist with pressure on household purchasing power. [Office for National Statistics · 2460] [Office for National Statistics · 2463] [Office for National Statistics · 2497]
The domestic weakness of the model is geographic concentration
London captures a disproportionate share of high-productivity and high-pay service activity. Output per hour in London was 28.5% above the UK average in 2023. GDHI per head in 2024 was £36,487 in London and £20,562 in the North East, while London also had a 40.8% high-pay job share in 2025. [Office for National Statistics · 2452] [Office for National Statistics · 2453] [Office for National Statistics · 2497] [Office for National Statistics · 2498] [Office for National Statistics · 2493]
This does not mean every region outside the South East is losing. The North West made a strong contribution to productivity growth from 2019 to 2023, and government strategy explicitly seeks to expand professional and business-service clusters outside London. The issue is that convergence in growth rates has not erased the level gap. [Office for National Statistics · 2452] [UK Government · 2440]
Counterevidence and alternative explanations
Brexit cannot explain the entire shift toward services because the structural change predates the 2016 referendum by decades. Automation, globalisation, changing consumption patterns and the 2008 financial crisis were already reshaping the industry mix before the new UK-EU relationship took effect. [Office for National Statistics · 2442] [OECD · 2448] [Office for National Statistics · 2429]
Post-Brexit growth in real other-business-services exports to the EU is also counterevidence to a blanket claim of services disintegration. At the same time, the financial-export decline and banking evidence reject the opposite claim that services were essentially untouched by leaving the single market. [Office for National Statistics · 2477] [Office for Budget Responsibility · 2454] [Bank of England · 2457]
Manufacturing’s lower share is not proof of permanent output collapse either. ONS revisions can materially change annual industry growth estimates, and some advanced manufacturing remains a policy priority. Sector share, real output, investment and employment need to be read together. [Office for National Statistics · 2418] [UK Government · 2440]
A two-axis service-growth matrix
Combining the evidence produces a more useful classification than a single services label. Digital and some professional services have strong export momentum and comparatively low dependence on passport-like authorisations. Finance has deep global advantages but much higher exposure to licensing, local-entity and regulatory-equivalence constraints. [Office for National Statistics · 2419] [Office for National Statistics · 2420] [Financial Conduct Authority · 2436] [Bank of England · 2457]
The matrix is not a statistical scorecard. It is a qualitative way to separate two questions that are often conflated: whether demand for the service is growing and whether the service can reach foreign customers without costly regulatory duplication. [HM Treasury · 2435] [HM Treasury · 2439] [HM Treasury · 2455]
Policy has moved toward a portfolio rather than pure services substitution
Recent UK policy does not assume that finance and digital services can make manufacturing irrelevant. The industrial strategy keeps advanced manufacturing as a growth sector while financial reforms, the UK-EU financial-services memorandum and the Berne agreement seek to widen access and competitiveness for services. [UK Government · 2440] [HM Treasury · 2455] [HM Treasury · 2431] [Office for Budget Responsibility · 2432] [HM Treasury · 2435] [HM Treasury · 2439]
That portfolio makes economic sense given the evidence. Services provide external surpluses and high-productivity activity; manufacturing provides supply-chain capacity, technology spillovers and geographically broader production. The practical policy question is how to reinforce complementarities rather than choose one side. [UK Government · 2440] [HM Treasury · 2455]
Scenarios and practical implications
In an upside scenario, professional and digital exports continue to grow, regulatory cooperation lowers the fixed cost of serving EU clients, and new regional clusters spread high-value employment. The UK could then preserve its services surplus while reducing the domestic concentration of its gains. [HM Treasury · 2435] [HM Treasury · 2431] [UK Government · 2440]
In a middle scenario, exports remain strong but aggregate productivity and financial access improve only slowly. The result could be robust trade performance without equally strong real-wage and regional-income growth – a continuation of the gap between external competitiveness and domestic living standards. [OECD · 2448] [Office for Budget Responsibility · 2432] [Office for National Statistics · 2429]
In a downside scenario, global financial stress or deeper regulatory fragmentation combines with weak industrial investment. The same international balance sheet that supports the City would transmit more external risk, while service surpluses would be unable to recreate lost local supply chains quickly. [Bank of England · 2430] [Bank of England · 2486] [Bank of England · 2487]
Limitations
The analysis deliberately avoids a single numerical “replacement rate” because the available indicators are not commensurable. GVA, cross-border services trade, FDI, bank claims, employment and household consumption answer different questions and belong to different accounting systems. [Bank of England · 2464] [Eurostat · 2465] [Office for National Statistics · 2463]
Nominal export values also contain price and exchange-rate effects. Where possible, real chained-volume series are used alongside current-price measures, but detailed digital services do not always have equally long real series. Regional outcomes also reflect housing costs, commuting, capital intensity and public investment as well as sector composition. [Office for National Statistics · 2420] [Bank of England · 2491] [Office for National Statistics · 2452]
Conclusion
What is known is that services became a powerful external stabiliser and that the export base broadened beyond finance. By 2025 the services surplus absorbed most of the goods deficit, while business and digital exports were far larger than at the start of the century. [Office for National Statistics · 2423] [Office for National Statistics · 2419] [Office for National Statistics · 2420]
What is conditional is the translation of that export strength into economy-wide productivity and living standards. Regulatory access, financial-sector performance and geographic diffusion determine whether a service-led model becomes broad-based growth or remains concentrated. [Bank of England · 2457] [Office for Budget Responsibility · 2454] [Office for National Statistics · 2452] [Office for National Statistics · 2453]
What cannot be answered with one defensible number is the percentage of manufacturing weakness “replaced” by services. The best evidence-based answer is asymmetric: the offset was very large in the trade balance, substantial in selected high-value activities, but incomplete and uneven in domestic production, jobs and regions. [Office for National Statistics · 2423] [Office for National Statistics · 2425] [Office for National Statistics · 2452]
Sources and method
The report cross-checks ONS national accounts, trade, jobs and regional-income data with Bank of England banking and stability statistics, OBR Brexit analysis, and UK/EU regulatory documents. Nominal and real series are kept distinct, and indicators from different accounting frameworks are compared conceptually rather than added together. [Office for National Statistics · 2442] [Office for National Statistics · 2423] [Bank of England · 2457] [Office for Budget Responsibility · 2454] [Bank of England · 2430]
| Indicator | Observed value | Interpretation |
|---|---|---|
| Manufacturing GVA share | 2024: 8.6% | Relative weight of manufacturing in a service-heavy economy |
| Telecom/computer/info exports | 2000 £5.459bn → 2025 £49.861bn | Long-run nominal digital-service expansion |
| Real other business services exports | 2000 £54.020bn → 2025 £183.340bn | Large real expansion of professional/business services |
| Services / goods balances | 2025 +£203.8bn / -£225.6bn | Services as a major external-balance buffer |
| Vulnerability | Official evidence | Implication |
|---|---|---|
| EU financial access | Passporting ended; affected EEA bank balances fell ~45% more | Regulation matters most for highly licensed services |
| Financial exports | OBR: 5.9% below 2019 | Finance and digital services do not share one growth path |
| Regional concentration | London productivity +28.5%; GDHI £36,487 vs £20,562 | Export success is not evenly distributed |
| Global financial exposure | Foreign claims measured in several trillions of dollars | Income opportunity and external-shock channel coexist |
How service exports translate into domestic productivity
Export demand can raise productivity when firms scale up, specialise and invest in skills or intangible capital, but the transmission is not automatic. If export growth is concentrated in a small number of established firms, the employment and supplier effects can remain narrow. If regional professional firms, universities and technology suppliers join the chain, the domestic spillover is broader. [Office for National Statistics · 2429] [UK Government · 2440] [Office for National Statistics · 2452]
The relevant comparison is therefore not services versus manufacturing in the abstract but the quality of the links between exporting firms and the domestic economy. Sectoral productivity, wages and regional income provide a better test of that transmission than export revenue alone. [Office for National Statistics · 2452] [Office for National Statistics · 2453] [Office for National Statistics · 2497]
Manufacturing and services are complements as well as substitutes
Modern manufacturing embeds design, software, finance, insurance, logistics, legal advice and research. A smaller manufacturing share and a larger services surplus are not necessarily a zero-sum shift. Strong advanced manufacturing can generate demand for professional services, while strong finance and digital capabilities can support manufacturers investment and overseas expansion. [UK Government · 2440] [HM Treasury · 2455]
This complementarity helps explain why current industrial strategy preserves advanced manufacturing as a growth sector instead of relying on services alone. The resilience question is whether the UK keeps high-value links across both parts of the economy, not which sector wins a simple size comparison. [UK Government · 2440] [HM Treasury · 2431]
Exchange rates and prices complicate the export story
Current-price exports reflect quantities, prices and exchange rates. Sterling depreciation can raise the sterling value of foreign-currency revenue without producing the same increase in real service volume, and contract currencies differ across activities. Bank of England research on the post-2016 shift away from sterling invoicing in some non-EU exports reinforces the need to avoid a simple weak-currency-equals-competitiveness rule. [Bank of England · 2491]
For that reason the report uses chained-volume measures where they are available and labels digital current-price series clearly. A durable service-led growth model should ultimately be judged by real volumes, productivity, wages and the breadth of domestic spillovers rather than nominal export values alone. [Office for National Statistics · 2420] [Office for National Statistics · 2419] [Office for National Statistics · 2429]
Green Map creates custom-edited map images using open geographic data sources such as geoBoundaries, Natural Earth, OpenStreetMap, and government open data. These maps are edited visual materials, not raw data files, and are provided for education, documents, presentations, and graphic reference.