86 local authorities projected minority White British by 2051

Correction, 12 October 2026. This piece said the cohort-component model gives 55.9% White British by 2051 on the same weighting and the same 312 areas, and that the two models agreeing to within half a point was a check on the recalibration. The cohort-component run had read each area’s ONS population projection twice, adding the single years of age to the all-ages total, so every population target was doubled. Re-run without that error it gives 54.7% on the same 312 areas. The Hamilton-Perry figure is unchanged at 56.3%, so the two models are now 1.6 points apart, and the half-point agreement this piece pointed to no longer holds. Nothing in the Hamilton-Perry count of 86, or the 59 majority-today areas, changes.

Correction, 7 September 2026. The national-picture paragraph was stale and named a denominator no subset of the dataset has. Recomputed with the function the homepage and national page use, the population-weighted White British share is 74.4% in 2021 across all 318 areas and 56.3% by 2051 across the 312 where 2051 is publishable, with 54.3% across the 263 areas the current model run produces and 55.9% for the cohort-component model on the same 312. The count of areas whose 2051 is withheld reads six, not fifteen.

Correction, 13 August 2026. This piece originally said 109 areas and cited the model’s backcast score as validation. Both were wrong. The backcast fitted its ratios on the same two Censuses it was tested against, so it measured the model’s own guardrails rather than its accuracy, and those guardrails were letting projections compound into arithmetic rather than demography.

The model has since been recalibrated against a genuine out-of-sample test that fits on Census 2001 to 2011 and forecasts 2021, scored on the actual Census 2021. That test showed the old settings were projecting change too fast, not too slow as previously stated here. On the recalibrated model the count is 86, and the currently-majority subset is 59, not 76. See the methodology for the test and the numbers.

86 councils. Minority White British by 2051.

67 by 2041. 86 by 2051. These are not forecasts. They are Hamilton-Perry projections extrapolating Census 2011-to-2021 cohort change ratios forward, constrained by ONS population projections. On the out-of-sample test the model is close to unbiased, with a mean error of +0.05 percentage points on the White British share and a mean absolute error of 1.56pp on that one-decade forecast. Read the numbers with that error attached rather than as point estimates.

Of the 86, 59 have a White British majority today. They cross the threshold within a generation.

The ten fastest transformations among current-majority areas:

Area	WBI 2021	WBI 2051	Change
Bolton	68.8%	26.9%	-41.9pp
Havering	66.5%	24.8%	-41.7pp
Pendle	66.1%	28.7%	-37.4pp
Thurrock	66.2%	29.3%	-36.8pp
Bexley	64.5%	28.5%	-36.0pp
Broxbourne	68.6%	33.1%	-35.5pp
Oldham	65.2%	29.9%	-35.3pp
Sandwell	52.1%	17.8%	-34.3pp
Dartford	67.3%	33.1%	-34.2pp
Sutton	57.3%	24.2%	-33.1pp

Bolton. 68.8% White British today. Projected 26.9% by 2051. That is a 42 percentage point decline in 30 years.

This is not a London story. Bolton, Pendle, Oldham, Thurrock, Broxbourne, Sandwell, Dartford. Towns across England.

Already below 50% in 2021 (33 areas): Birmingham (42.9%), Leicester (33.2%), Luton (31.8%), Slough (24.0%), Newham (14.8%), Brent (15.2%). These areas passed the threshold years ago.

The national picture. England and Wales White British share: 74.4% (Census 2021, ONS TS021). This site’s own weighting gives the same 74.4% across all 318 areas in the dataset, and 56.3% by 2051 across the 312 areas where 2051 is publishable, which is the figure the homepage carries. The 2051 share is not all one model: 49 of those areas carry projections from an earlier run the current code cannot reproduce, 15.4% of the weighted population, and across the 263 areas the current model produces 2051 is 54.3%. The cohort-component model, which assumes ethnic fertility rates converge partway to the national mean, gives 54.7% on the same weighting and the same 312 areas.

The two models are 1.6 points apart on this weighting, 56.3% against 54.7%. This piece previously said they agreed to within half a point and called that a check on the recalibration; that agreement came from a cohort-component run with an error in it, and it does not hold on the corrected run.

Three things drive it. Fertility: White British TFR 1.31, below replacement. Age structure: the White British population is older, dying faster than it replaces itself. Migration: selective out-migration of White British to surrounding areas, combined with international in-migration to cities.

Why a handful of areas are still missing from this count. Cohort change ratios compound, and with the old growth ceiling of 5.0 a group could quintuple in a decade, which is 625 times over four steps. That produced projections that were arithmetic rather than demographic in 108 areas: Enfield’s “Other” category reached 67% by 2051, Barnsley’s White Other went from 4.3% to 45.3%. The ceiling selected on the out-of-sample test cuts that to six areas whose 2051 is still withheld: Barnet, Hackney, Lambeth, Enfield, Haringey and Islington. Those six do not appear in the 86.

The data is on every place page on this site. Search your area.

Source: Census 2021 custom dataset (20 ethnic groups, direct observations). Hamilton-Perry v8.0 single-year CCR model, Census 2011 DC2101EW base, SNPP 2022-based envelope. Ratios shrunk toward the national ratio by cell size and capped at 1.65 growth per decade, both selected on an out-of-sample test fitting 2001 to 2011 and forecasting 2021 (MAE 1.56pp, bias +0.05pp across 285 areas).