The New Business Map: UK
Where Britain is starting new businesses since 2019, and whether local incomes kept up.
Headline finding
New businesses per adult fell in 291 of 336 UK councils between 2019 and 2024. They fell least where unemployment claims fell and household incomes grew.
Britain is registering more companies than before the pandemic, but fewer of them become trading businesses. ONS business births, which count businesses that register for VAT or PAYE, fell 16.6% per adult across the UK, from 6.9 to 5.8 per 1,000 adults. The fall reached almost every part of the country, but it was gentler where local economies held up: where the claimant rate fell, the typical council lost 8.7% of its new-business rate; where claims rose by more than a point, it lost 15.7%.
The typical council is the median. Claims are the claimant count as a share of residents aged 16 to 64, 2019 against 2025. Source: Censal Research analysis of ONS data.
Key stats
- UK business births fell 12.7% between 2019 and 2024, from 363,825 to 317,440; per adult, they fell 16.6%, from 6.9 to 5.8 per 1,000 adults, according to Censal Research's analysis of ONS data.
- Business births per adult fell in 291 of the 336 UK councils with at least 50,000 adults between 2019 and 2024. The median council fell 12.6%.
- Companies House incorporations rose 20.5% over the same period, from 665,495 in 2019-20 to 801,864 in 2024-25, while the number of new VAT- or PAYE-registered businesses fell.
- Where the claimant rate fell, business births per adult fell a median 8.7%; where it rose by more than one percentage point, they fell 15.7% (claimant rate 2019 against 2025).
- Where real household income per head fell by more than 2.5%, business births per adult fell a median 16.6%; where it rose by more than 5%, they fell 7.9% (GDHI per head, 2019 to 2024, adjusted for inflation).
- London outside the centre had the steepest fall of any UK nation or region (median council −20.5%); Northern Ireland had the smallest (−5.3%).
- Of the 15 UK councils with more than 400,000 adults, Bradford is the only one where business births per adult rose (+9%); Bristol fell most (−25%).
- 165 UK postcodes are each the registered address of at least 1,000 live companies, 652,645 companies in all, about one in nine on the register.
Source: Censal Research, The New Business Map: UK (2026), analysis of ONS Business Demography, ONS regional household income, ONS claimant count and Companies House data.
1. Fewer new businesses almost everywhere
Of the 336 councils we rank (at least 50,000 adults, 96.1% of UK adults), 291 recorded fewer business births per adult in 2024 than in 2019. The median council fell 12.6%. England and Wales fell furthest (median council −13.2% in both), Scotland −10.9% and Northern Ireland least (−5.3%). Within England, London outside the centre fell most (−20.5%) and the North East (−7.3%) and West Midlands (−7.5%) least.
Median of ranked councils. Central London councils are not ranked (see "How we handled the tricky parts"). Source: Censal Research analysis of ONS data.
2. Births held up where local economies held up
The fall was steepest where the local labour market weakened. In the 48 councils where the claimant rate fell between 2019 and 2025, the median council's births per adult fell 8.7%. In the 131 councils where it rose by more than one percentage point, they fell 15.7%. Councils in between sit in between.
Household income tells the same story at the ends of the range. Where real disposable household income per head fell by more than 2.5%, births per adult fell a median 16.6%; where it rose by more than 5%, they fell 7.9%. Across all councils the correlation between income change and births change is 0.21 (95% interval 0.10 to 0.31); for the claimant rate it is −0.12 (−0.21 to −0.02). Both are modest, and both point the same way: this was not a wave of people starting businesses because local jobs or incomes dried up.
The pattern survives a stricter test. Using three-year averages (2017–19 against 2022–24), which smooth out one-off spikes, births per adult fell in 270 of 344 councils; the median fall was 5.4% where claims fell and 13.8% where they rose by more than a point, and 14.5% where incomes fell by more than 2.5% against 6.1% where they rose by more than 5%.
Household income is gross disposable household income (GDHI) per head, adjusted for inflation with CPIH. The pattern is an association, not proof of cause. Source: Censal Research analysis of ONS data.
3. Big cities: Bradford rose, Bristol and Liverpool fell hardest
Among the 15 ranked councils with more than 400,000 adults, business births per adult fell most in Bristol (−25%), Liverpool (−23%), Manchester (−20%) and Glasgow (−19%). Bradford is the only one where they rose, from 5.5 to 6.0 per 1,000 adults (+9%, and +12% on three-year averages). Leeds (−8%) and Sheffield (−7%) held up better than the typical council.
Source: Censal Research analysis of ONS data.
Show the numbers as a table
| Council | Adults (2024) | Births per 1,000 adults, 2019 | 2024 | Change | Change, 3-year averages |
|---|---|---|---|---|---|
| Bradford | 419,482 | 5.5 | 6.0 | +9.1% | +11.8% |
| Sheffield | 466,056 | 5.0 | 4.6 | −7.0% | −0.2% |
| Cornwall | 476,737 | 4.8 | 4.4 | −7.0% | +0.7% |
| County Durham | 436,928 | 3.9 | 3.6 | −8.1% | +0.6% |
| Leeds | 669,595 | 6.1 | 5.6 | −8.4% | +0.6% |
| North Yorkshire | 518,799 | 5.1 | 4.6 | −9.4% | −5.5% |
| City of Edinburgh | 447,174 | 6.2 | 5.4 | −12.5% | −15.4% |
| Somerset | 476,358 | 5.1 | 4.4 | −13.9% | −0.7% |
| Buckinghamshire | 448,942 | 8.0 | 6.8 | −14.0% | −13.8% |
| Birmingham | 895,487 | 8.7 | 7.3 | −16.1% | −17.4% |
| Wiltshire | 420,448 | 5.3 | 4.3 | −17.6% | −24.7% |
| Glasgow City | 540,490 | 6.5 | 5.3 | −19.3% | −11.6% |
| Manchester | 460,233 | 10.4 | 8.3 | −20.2% | −33.5% |
| Liverpool | 412,383 | 7.4 | 5.7 | −23.1% | −19.7% |
| Bristol, City of | 403,414 | 7.2 | 5.4 | −25.0% | −13.8% |
4. More companies, fewer businesses
Companies House incorporations rose from 665,495 in 2019-20 to 801,864 in 2024-25 (+20.5%), and 815,277 in 2025-26. Over the same years ONS business births fell 12.7%. The two measures count different things. A company can be formed online in a day and left dormant, used to hold property or investments, or run below the VAT threshold with no staff. A business birth in ONS data means a business has registered for VAT or PAYE: it is trading at scale or employing someone.
So the paperwork boom is real, but it has not turned into more trading businesses. Company registrations also tell you little about place: 165 postcodes are each the registered address of at least 1,000 live companies, 652,645 companies in all, about one in nine on the register. The largest single postcode holds 88,436. That is why this study ranks councils on ONS births, not on incorporations.
Sources: Companies House, Companies register activities, Table A8; ONS, Business demography, UK.
5. Self-employment shrank; pay data are too thin to say more
The self-employed share of people in work fell in the typical council from 14.5% in 2019 to 12.9% in 2025 (Annual Population Survey). The fall is statistically clear in 38 councils, and a rise is clear in 3 (95% level, before correcting for testing every council). Changes in self-employment are unrelated to changes in business births (correlation 0.04).
We also tested full-time median pay by council (Annual Survey of Hours and Earnings). Real pay rose a median 0.7% between 2019 and 2025, but council estimates are too imprecise to test: no council's real pay change is statistically clear. We therefore use household income, which ONS models for every council, as the income measure.
Where births rose fastest and fell furthest, and why single years need care
Rankings at the top and bottom move a lot from one year to the next. Derry City and Strabane had the biggest single-year rise, but on three-year averages its gain is +11%. Pendle, Torfaen and Erewash rose on both measures. At the other end, Test Valley, Eastleigh and Brighton and Hove fell sharply on both. Read the three-year column alongside any individual rank.
| Council | Region | Births per 1,000 adults, 2019 → 2024 | Change | Change, 3-year averages |
|---|---|---|---|---|
| Derry City and Strabane | Northern Ireland | 3.7 → 6.5 | +73.1% | +10.8% |
| Pendle | North West | 4.5 → 5.5 | +23.0% | +30.3% |
| Torfaen | Wales | 3.6 → 4.3 | +21.5% | +52.1% |
| West Suffolk | East of England | 5.2 → 6.2 | +19.4% | +12.6% |
| South Kesteven | East Midlands | 5.4 → 6.3 | +16.4% | +1.9% |
| Council | Region | Births per 1,000 adults, 2019 → 2024 | Change | Change, 3-year averages |
|---|---|---|---|---|
| Test Valley | South East | 10.3 → 4.4 | −57.1% | −43.0% |
| Eastleigh | South East | 9.9 → 4.8 | −52.1% | −59.9% |
| Brighton and Hove | South East | 11.8 → 6.3 | −46.6% | −32.0% |
| Watford | East of England | 13.3 → 7.8 | −41.7% | −23.6% |
| Fife | Scotland | 5.0 → 3.0 | −40.6% | −28.6% |
The three-year column compares 2017–19 with 2022–24. Where the two columns disagree sharply, one unusual year is driving the single-year change.
Two countries, one local pattern
Our first study, The New Business Map, found that US business applications per adult rose 53% between 2019 and 2025, and grew faster where household incomes rose. Britain looks different on the surface: trading-business births fell. But on paperwork measures both countries boomed (US applications +62%, UK incorporations +20%), and in both, new businesses did better where local incomes did better.
The measures are not the same. US applications count requests for a tax number, a signal of intent; UK births count businesses that have registered for VAT or PAYE. Compare the direction of each country's local pattern, not the levels.
How we handled the tricky parts
Central London
Not ranked. In the City, Westminster, Camden, Islington and Hackney, head offices and registered offices, not residents, drive the counts, so a rate per resident adult would mislead. Bromsgrove is set aside on the same rule.
An inflated 2019
Eight councils had 2019 births more than 1.4 times the higher of their 2017 and 2018 levels, which would exaggerate any fall. They are not ranked, and every result is rechecked on three-year averages.
Council mergers
Every year is rebuilt on today's 361 councils using the official reorganisation lists. For every year from 2017 to 2024, council totals add up exactly to the UK total.
Pay against household income
Council pay surveys are too small to measure change reliably, so we use disposable household income per head, adjusted for inflation, which ONS publishes for every council.
Registration mailboxes
We located every live company with the Companies House register and flagged councils where mass-registration addresses could distort births. Tower Hamlets is ranked but flagged as a caution.
Companies against businesses
We rank councils on ONS business births, which need a VAT or PAYE registration, rather than on company registrations, which are cheap, often dormant and filed at mailbox addresses.
How we did it, in plain English
Read the full methodologyWe took ONS business births for every UK council in 2019 and 2024 (Business demography, UK, latest edition) and divided by the adult population (ONS mid-year estimates). We compared the change with the change in real disposable household income per head (ONS, 2019 to 2024, adjusted with CPIH), the claimant count rate (2019 against 2025) and the self-employed share of workers (Annual Population Survey), and set it against Companies House registrations.
Councils with fewer than 50,000 adults were excluded to keep rates stable, and the screens above were applied by published rule. Survey-based measures were tested against their published confidence intervals. Every headline number was recomputed by a second, independently written program.
Limitations
- Associational only. This compares two periods across councils. It cannot say that incomes or claims cause changes in business births, or the reverse.
- Births miss the smallest firms. Businesses below the VAT threshold with no employees are not counted. The threshold rose from £85,000 to £90,000 in April 2024, which may lower 2024 births slightly.
- Location by registered address. Births are placed where the business is registered, which can differ from where it trades.
- Modelled income. Household income by council is modelled by ONS and has no published margin of error; it can be revised.
- Different years. Births and income run to 2024, the claimant count and self-employment to 2025.
Questions people ask
Are fewer new businesses being started in the UK?
Yes, on the official measure. ONS business births, which count businesses registering for VAT or PAYE for the first time, fell from 363,825 in 2019 to 317,440 in 2024, a 12.7% fall. Per adult, births fell 16.6%, from 6.9 to 5.8 per 1,000 adults. They fell in 291 of the 336 UK councils Censal Research ranked.
Why are company registrations rising when new businesses are falling?
Because registering a company and starting a trading business are different things. Companies House incorporations rose from 665,495 in 2019-20 to 801,864 in 2024-25 (+20.5%). Many new companies are dormant, hold property or investments, or never pass the VAT threshold or employ anyone, so they never appear as ONS business births.
Which UK cities have seen the biggest fall in new businesses since 2019?
Among councils with more than 400,000 adults, business births per adult fell most in Bristol (−25%), Liverpool (−23%), Manchester (−20%), Glasgow (−19%), Wiltshire (−18%) and Birmingham (−16%) between 2019 and 2024. Bradford was the only one where they rose (+9%).
Where in the UK are new businesses holding up best?
Northern Ireland (median council −5.3%), the North East (−7.3%) and the West Midlands (−7.5%) saw the smallest falls in business births per adult between 2019 and 2024. London outside the centre saw the largest (−20.5%). Pendle, Torfaen and Erewash are among the few councils where births per adult rose on both a single-year and a three-year comparison.
Is the fall in new businesses linked to unemployment?
Locally, yes, modestly. In the 48 councils where the claimant rate fell between 2019 and 2025, business births per adult fell a median 8.7%. In the 131 where it rose by more than one percentage point, they fell 15.7%. The correlation is weak (−0.12), and the comparison is descriptive, not causal.
Did household incomes affect new business formation in the UK?
Births held up better where incomes grew. Where real disposable household income per head fell by more than 2.5% between 2019 and 2024 (55 councils), business births per adult fell a median 16.6%. Where it rose by more than 5% (26 councils), they fell 7.9%. The correlation across councils is 0.21.
Why is central London left out of the rankings?
In the City of London, Westminster, Camden, Islington and Hackney, business births per adult are at least three times the UK rate, because head offices and registered offices are counted where they are registered, not where their owners live. A rate per resident adult would be misleading, so these councils appear in the data table but are not ranked.
What counts as a business birth in UK statistics?
The Office for National Statistics counts a business birth when an enterprise registers for VAT or PAYE for the first time and was not active in the previous two years. Businesses below the VAT threshold with no employees are not counted, which makes births a stricter measure than company registrations.
How to cite this study
Censal Research (2026). The New Business Map: UK. Where Britain is starting new businesses since 2019, and whether local incomes kept up. Version 1.0, October 2026. https://censalresearch.com/studies/new-business-map-uk/
Short form for journalists: "according to an analysis of ONS data by Censal Research".
Derived tables are published under CC BY 4.0; please credit "Censal Research, The New Business Map: UK (2026)".
Data
- Council table (CSV, 361 rows)Every UK council: business births and births per adult in 2019 and 2024, changes on single-year and three-year bases, household income, claimant and self-employment measures, company registrations, exclusion reasons and flags.
- MethodologySources, definitions, tests, boundary changes, exclusion rules and the corrections log.
Journalists and researchers can ask for the analysis code or a cut of the data for their area at [email protected].
Contains public sector information licensed under the Open Government Licence v3.0. Source: Office for National Statistics and Companies House. Contains OS data © Crown copyright and database right; contains Royal Mail data © Royal Mail copyright and database right. Censal Research is not affiliated with or endorsed by ONS or Companies House.
Author: Censal Research. Contact: [email protected]. Corrections policy: errors are corrected in place with a dated note.
About this edition: v1.0 (October 2026) uses business births to 2024, the latest ONS has published. We will update this study when newer official data is published.