See who runs things in Llanelli
Map of Llanelli
The map needs JavaScript. The figures below do not.
- Llanelli constituency
- The council area it sits in (approximate outline)
- Marker: a point inside the constituency
The outline is the constituency. Boundaries: Office for National Statistics, Open Government Licence. Use Full screen for the layers: neighbourhood deprivation (IoD2025, England only) and nearby schools. Map data © OpenStreetMap contributors.
95,434
Resident population · 2024
0.6%
Population change · 2022 to 2024
1.1 percentage points lower than WalesCarmarthenshire 0.9%Wales 1.7%
How do we know this?
Method: Mid-year population estimates
ONS mid-year estimates are modelled from the 2021 Census, births, deaths and migration. Mid-2022 is the revised series. Ages are grouped: 0-15 means ages 0 to 15 inclusive, working age 16-64, 65 and over. Change 2022-2024 compares the revised mid-2022 with mid-2024 estimates for the same boundaries. Scotland and Northern Ireland: mid-year estimates for council areas published by ONS from National Records of Scotland and NISRA (via Nomis), for 2022 to 2024, to match the England and Wales years. No small-area (ward or neighbourhood) estimates are held for them.
£179,995
Median sale price · Year to Oct 2025
16% lower than WalesCarmarthenshire £193,000Wales £215,000
5.9%
Change in median sale price · Year to Oct 2025 vs a year earlier
3.5 percentage points higher than WalesCarmarthenshire 4.4%Wales 2.4%
How do we know this?
Method: Price paid medians
Sales are mapped from postcode to each geography exactly. The headline is the median sale price over the latest 12 COMPLETE months (a month is complete when national standard sales are at least 85% of the average of the 24 months before it; currently November 2024 to October 2025) and the 12 months before that. Months after that (Nov 2025 onward in the April 2026 extract) are still being registered and are excluded. Yearly medians use complete calendar years 2016-2024. Medians need at least 10 sales; sales counts are published for any area. Medians describe homes that sold, not all homes, and are not adjusted for the mix of property types, so a change in the median can reflect what was sold. Registration lags completion, so recent months can be revised up. Only standard price paid sales (category A) are included.
31.3%
Adults with a degree-level qualification or above · 2021
Carmarthenshire figure
About the same as WalesWales 31.5%
20%
Adults with no qualifications · 2021
Carmarthenshire figure
About the same as WalesWales 19.9%
How do we know this?
Method: Census 2021 housing, health and qualifications for Wales
Wales held its census on the same day as England (21 March 2021) with the same questions, so the figures are directly comparable. For Wales we use the published local authority tables. Tenure counts shared ownership with owners; "private rented" includes people who live rent free, as in the England figures. Overcrowding is an occupancy rating of -1 or less for bedrooms. Qualifications are for people aged 16 and over; "no qualifications" and "Level 4 or above" (degree level and above) are shares of all those people. Counts are rounded by random adjustment, so tiny differences from other tables are expected. Wales households per local authority are not split to small areas here.
3,314
Companies registered in the area · 2 Oct 2026
34.7 per 1,000
Registered companies per 1,000 residents · 2 Oct 2026
37% lower than WalesCarmarthenshire 43.4 per 1,000Wales 55.2 per 1,000
How do we know this?
Method: Registered companies by area
Companies are placed by the postcode of their REGISTERED OFFICE, which is not a trading or operating location: accountants, formation agents and virtual offices register thousands of companies at one address, and many small companies use a director's home. Counts are of companies with status Active (including those with a proposal to strike off), by the section of their first listed SIC 2007 code. Incorporation counts cover companies incorporated in the window that are still on the register at 2 Oct 2026, so dissolved companies are missing and older windows are understated relative to newer ones. Companies per 1,000 residents divides by the mid-2024 resident population (England and Wales only) and is not a measure of business density where registration hubs exist. SIC section Z groups dormant, non-trading and unclassified companies.
How much of the work residents do generally available AI tools could speed up Carmarthenshire figure
60.1%17.8%20.7%
- A small part 60.1%
- Some 17.8%
- Much 20.7%
- Not yet assessed 1.4%
Early estimate from CareerMash (October 2026). It may change and has not been independently checked.
What stands out about this place?
Findings where this place differs most from its surroundings, with the evidence behind each one. A difference is not a verdict: it is a prompt to look closer.
HomesHigh confidence
Change in the typical home sale price over a year: higher than Wales
Change in the typical home sale price over a year is 5.9% in Llanelli (Year to Oct 2025 vs Year to Oct 2024), compared with 2.4% in Wales (3.5 percentage points higher).
5.9% here2.4% Wales
How do we know this?
- Source
- HM Land Registry: Price Paid Data (standard price paid, category A) (Open Government Licence v3.0 (HM Land Registry Price Paid Data, Crown copyright))
- Period
- Year to Oct 2025 vs Year to Oct 2024
- Geography
- Constituency, compared with Wales
- Method
- Price paid medians. Sales are mapped from postcode to each geography exactly. The headline is the median sale price over the latest 12 COMPLETE months (a month is complete when national standard sales are at least 85% of the average of the 24 months before it; currently November 2024 to October 2025) and the 12 months before that. Months after that (Nov 2025 onward in the April 2026 extract) are still being registered and are excluded. Yearly medians use complete calendar years 2016-2024. Medians need at least 10 sales; sales counts are published for any area. Medians describe homes that sold, not all homes, and are not adjusted for the mix of property types, so a change in the median can reflect what was sold. Registration lags completion, so recent months can be revised up. Only standard price paid sales (category A) are included.
- Confidence
- High confidence: the figure rests on large numbers and good-quality data.
- Limitations
- Covers the 12-month window Year to Oct 2025 vs Year to Oct 2024 only.
EconomyMedium confidence
Change in dissolutions on the prior 12 months: higher than Wales
Change in dissolutions on the prior 12 months is -2.7% in Llanelli (Year to 13 Nov 2026 vs Year to 13 Nov 2025), compared with -29.7% in Wales (27.0 percentage points higher).
-2.7% here-29.7% Wales
How do we know this?
- Source
- Companies House: Dissolved companies: advanced search API (one query per dissolution day since 2010), topped up from the live streaming feed; incorporations from the register snapshot of 2 Oct 2026 (to confirm (Companies House API terms; the register data is published under the Open Government Licence v3.0))
- Period
- Year to 13 Nov 2026 vs Year to 13 Nov 2025
- Geography
- Constituency, compared with Wales
- Method
- Company dissolutions and net formation. Dissolutions are companies Companies House records as dissolved, counted on the dissolution date and located by the postcode of the registered office held at dissolution. That address is often an accountant, formation agent or liquidator, not where the company traded, so it says where a company was registered, not where a business closed. Every dissolution day since 1 January 2010 was fetched one day at a time and checked: the number of companies stored equals the number the register reports for that day. The latest window is the 365 days ending the latest complete day; the prior window is the 365 days before it; annual figures are complete calendar years. The route to dissolution (strike-off, end of liquidation, merger) is not recorded, so dissolutions are not split. Net formation = companies incorporated in the period (all that were ever registered in it, whether or not they have since been dissolved) minus companies dissolved in the period. The two groups are different companies and are located by different addresses, current for companies still on the register and last held for dissolved ones. Net formation compares registered offices, not jobs, trading or business closures.
- Confidence
- Medium confidence: the figure is an estimate or rests on a moderate number of records, so read small differences with care.
- Limitations
- Registered offices are not operating locations; many companies are registered at an agent or accountant address.
- Registered-office events are not new local employers, closures or job changes.
The futureHigh confidence
Residents retired
27.8% of residents aged 16+ in Carmarthenshire (the council area containing Llanelli) were retired on Census day, higher than the 21.6% for England and Wales (Census 2021, during the COVID-19 pandemic).
27.8% here21.6% England and Wales
How do we know this?
- Source
- not recorded
- Period
- Census 2021 (21 March 2021)
- Geography
- Constituency, compared with England and Wales
- Method
- Compared with the same measure for the comparison geography.
- Confidence
- High confidence: the figure rests on large numbers and good-quality data.
- Limitations
- Census day was 21 March 2021, during the pandemic and furlough.
- Shown separately from other measures; no composite.
HomesMedium confidence
Wide range between wards in Llanelli: Share of home sales that are new-build
Share of home sales that are new-build (Year to Oct 2025) varies within Llanelli: from 0.0% in Bigyn to 32.9% in Pembrey, across 17 wards. The upper tenth of wards is 19.1 percentage points above the lower tenth.
Pembrey figure
32.9% Pembrey0% Bigyn
How do we know this?
- Source
- HM Land Registry: Price Paid Data (standard price paid, category A) (Open Government Licence v3.0 (HM Land Registry Price Paid Data, Crown copyright))
- Period
- Year to Oct 2025
- Geography
- Ward, compared with lowest-value ward (Bigyn)
- Method
- Price paid medians. Sales are mapped from postcode to each geography exactly. The headline is the median sale price over the latest 12 COMPLETE months (a month is complete when national standard sales are at least 85% of the average of the 24 months before it; currently November 2024 to October 2025) and the 12 months before that. Months after that (Nov 2025 onward in the April 2026 extract) are still being registered and are excluded. Yearly medians use complete calendar years 2016-2024. Medians need at least 10 sales; sales counts are published for any area. Medians describe homes that sold, not all homes, and are not adjusted for the mix of property types, so a change in the median can reflect what was sold. Registration lags completion, so recent months can be revised up. Only standard price paid sales (category A) are included.
- Confidence
- Medium confidence: the figure is an estimate or rests on a moderate number of records, so read small differences with care.
- Limitations
- Ward and neighbourhood boundaries differ in size; the range shows variation within the area, not a ranking of places.
EconomyMedium confidence
Increase between 2022 and 2025: Median weekly pay of full-time employees who live in the area (Carmarthenshire figure)
Median weekly pay of full-time employees who live in the area in Carmarthenshire council area went from £618 (2022) to £733 (2025): up 19% (not adjusted for inflation), compared with up 19.0% in Wales.
Carmarthenshire figure
£733 Carmarthenshire figure£618 same place, 2022
How do we know this?
- Source
- Office for National Statistics: Annual Survey of Hours and Earnings (ASHE), via Nomis: tables 7 (place of work) and 8 (place of residence) (Open Government Licence v3.0)
- Period
- 2025
- Geography
- Council area, compared with same place, 2022
- Method
- Earnings from the Annual Survey of Hours and Earnings. ASHE is a 1% sample of employee jobs from HMRC PAYE records, with a reference week in April. We hold the median gross pay of full-time employees, by place of residence (where the employee lives) and by place of work (where the job is), for the latest survey year and the years back to 2015. Median weekly pay is gross pay before tax for the pay period that includes the reference date; annual pay is gross annual pay for the year to April. Pay is in cash terms: it is not adjusted for inflation, so changes over time include price rises. Survey estimates have a margin of error that grows as areas get smaller; each value carries its coefficient of variation and values with a CV above 10% are marked low quality. Estimates ONS judges too unreliable (CV above about 40%) are not published and appear here as absent. Self-employed people are not covered. Residence-based figures describe people who live in the area wherever they work; workplace-based figures describe jobs located in the area wherever the employee lives. The two can differ a lot in commuter areas and city centres.
- Confidence
- Medium confidence: the figure is an estimate or rests on a moderate number of records, so read small differences with care.
- Limitations
- Coefficient of variation 4.3%. Coefficient of variation 5.9%.
- This figure is for the Carmarthenshire council area, not for Llanelli itself.
- Prices are nominal and not adjusted for inflation or for changes in the mix of homes sold.
The futureMedium confidence
Occupations where AI tools could speed up a small part of the work
An occupation-adjusted estimate puts 60.1% of employed residents of Carmarthenshire (the council area containing Llanelli) in occupations where AI tools could already speed up a small part of the work, higher than the 50.8% estimated for England and Wales by the same method. Across all employed residents: 60.1% in occupations where AI tools could speed up a small part of the work, 17.8% some, 20.7% much, and 1.4% not yet assessed. Based on Census 2021 occupations and the national mix of CareerMash categories inside each occupation group. This is about the tasks in the job, not whether there will be jobs. It does not predict job losses. Early estimate from CareerMash (October 2026). It may change and has not been independently checked.
60.1% here50.8% England and Wales
How do we know this?
- Source
- not recorded
- Period
- Census 2021 (21 March 2021)
- Geography
- Constituency, compared with England and Wales
- Method
- Compared with the same measure for the comparison geography.
- Confidence
- Medium confidence: the figure is an estimate or rests on a moderate number of records, so read small differences with care.
- Limitations
- Estimate for residents, not a count of jobs in the area.
- Resident basis: this counts people where they live, not where they work. Some residents work in other areas.
- Census occupations are as at 21 March 2021, a COVID-19 period. The CareerMash categories are task evidence dated 11 October 2026. This is not a description of current employment.
- Each local figure is an estimate: local occupation shares multiplied by a national mix inside each occupation group. It does not use local employers or local jobs, and a place can differ from the national mix inside a group.
- Early estimate from CareerMash (October 2026). It may change and has not been independently checked.
The constituency itself has no dated records, so these are from the wider council area. They are events, not judgements.
No records for this constituency: council area shown
- 13 Nov 2026
- 13 Nov 2026
- 13 Nov 2026
- 13 Nov 2026
Ask a question about Llanelli
Answers are worked out from the figures on these pages, not written by a chatbot, and each one lists its evidence.
Sources used on this page
9- Charity Commission for England and Wales; OSCR (Scotland); Charity Commission for Northern Ireland: Registers of charities (registered address, latest income, charitable purposes)Period: CCEW extract 4 Oct 2026; OSCR 7 Oct 2026 · Published: 4 Oct 2026 · Coverage: United Kingdom (registered charities) · Licence: Open Government Licence v3.0 (Charity Commission, OSCR and CCNI terms apply: not the register itself, no direct marketing)
- Police forces of England and Wales via data.police.uk: Street-level crime (12 monthly files)Period: May 2025 to Apr 2026 · Published: Jun 2026 · Coverage: England and Wales except Greater Manchester Police (no data in the loaded files); Northern Ireland records have no LSOA · Licence: Open Government Licence v3.0
- Companies House: Free Company Data Product (basic company data), snapshotPeriod: snapshot 2 Oct 2026 · Published: 2 Oct 2026 · Coverage: UK companies on the register (registered office postcode) · Licence: Open Government Licence v3.0 (Free Company Data Product)
- HM Land Registry: Price Paid Data (standard price paid, category A)Period: sales to 31 Mar 2026 · Published: Apr 2026 · Coverage: England and Wales · Licence: Open Government Licence v3.0 (HM Land Registry Price Paid Data, Crown copyright)
- Office for National Statistics: Lower layer Super Output Area mid-year population estimates by single year of age (mid-2022 revised, mid-2023, mid-2024)Period: mid-2022 (revised), mid-2023, mid-2024 · Published: 7 Nov 2025 · Coverage: England and Wales (LSOA 2021) · Licence: Open Government Licence v3.0
- Office for National Statistics: Annual Survey of Hours and Earnings (ASHE), via Nomis: tables 7 (place of work) and 8 (place of residence)Period: survey year 2025 (April 2025 reference week); 2015 to 2025 held · Published: Oct 2025 · Coverage: Great Britain local authorities (April 2023 boundaries), regions and countries; Northern Ireland not held · Licence: Open Government Licence v3.0
- Office for National Statistics: Census 2021 bulk tables via Nomis: TS054 tenure, TS045 cars, TS044 accommodation type, TS017 household size, TS052 occupancy rating (bedrooms), TS037 general health, TS067 highest qualificationPeriod: 21 March 2021 · Published: 2023 · Coverage: Local authorities of Wales (22) and Wales; qualifications also for England and its local authorities · Licence: Open Government Licence v3.0
- Companies House: Companies House streaming API (live), combined with the basic company data snapshot of 2 Oct 2026Period: incorporations to 13 Nov 2026 (stream timepoint 121711952, newest event published 14 Nov 2026T03:43:17) · Published: 14 Nov 2026 · Coverage: UK companies (registered office postcode)
- Companies House: Dissolved companies: advanced search API (one query per dissolution day since 2010), topped up from the live streaming feed; incorporations from the register snapshot of 2 Oct 2026Period: dissolutions 14 Nov 2024 to 13 Nov 2026; annual series 2010 to 2025 · Published: 14 Nov 2026 · Coverage: UK companies (registered office postcode)
Registered offices are not operating locations. Ranks and deciles are never averaged. Census 2021 figures are labelled 2021.