Blackburn projected minority White British by 2027. Three more Lancashire towns follow by 2050.

Correction, 12 October 2026. Blackburn with Darwen’s White British share in 2011 was 66.5% (Census 2011 KS201EW: 98,144 of 147,489 usual residents), not 66.6%. Preston’s asylum figure, 554 people on asylum support, could not be traced to any Home Office quarter: Preston had 429 people on asylum support at 31 December 2025, the quarter Burnley’s 461 comes from, and 407 at 30 June 2026. Burnley had 471 at 30 June 2026 (Home Office, Immigration system statistics, regional and local authority data). Nothing else in the piece changes.

Blackburn. 56.9% White British. About a year from minority status.

Census 2021: Blackburn with Darwen, 56.9% White British. Down from 66.5% in 2011. Nearly 10 percentage points gone in one decade.

The Hamilton-Perry model projects (v8.0, scored out of sample against Census 2021: mean absolute error 1.56pp on that one-decade forecast, bias +0.05pp across 285 areas):

Year	Blackburn WBI
2021	56.9%
2027	~50% (threshold)
2031	45.8%
2041	34.6%
2051	24.6%

Blackburn is not alone in Lancashire.

Pendle. 66.1% White British (2021). Projected below 50% by approximately 2034. Nelson and Colne are the concentration points. White British projected 28.7% by 2051.

Preston. 66.1% White British (2021). Projected below 50% by approximately 2037. University city. International migration pipeline. 429 on asylum support at 31 December 2025, 407 at 30 June 2026. White British projected 35.6% by 2051.

Burnley. 77.9% White British (2021). Projected below 50% by approximately 2050. 461 on asylum support at 31 December 2025, 471 at 30 June 2026. The Daneshouse and Stoneyholme ward is already majority South Asian.

Four Lancashire mill towns. Same industry. Same deprivation. Same trajectory. Different timelines.

Blackburn gets there first because it started from a lower base (56.9% vs Burnley’s 77.9%) and has a larger established South Asian community driving higher cohort change ratios. The dynamics are identical. Only the speed differs.

Source: Census 2021 custom dataset. Hamilton-Perry v8.0, Census 2011 DC2101EW base, SNPP-constrained, recalibrated on an out-of-sample test (MAE 1.56pp, bias +0.05pp). Threshold years revised on 13 August 2026. Threshold years interpolated from decadal projections. The backcast score previously quoted here has been withdrawn; it measured the model’s own guardrails rather than its accuracy. See the methodology.