Estimating Ethnic Populations between Censuses

Philip H. Rees , University of Leeds
Paul Norman, University of Leeds
Gabriele Price, Office of Health Improvement and Disparities, Department of Health and Social Care
Stephanie Webb, Office of Health Improvement and Disparities, Department of Health and Social Care

Topic. An ethnically diverse society committed to equality goals needs annual estimates of ethnic group populations to compute key welfare indicators. Theoretical focus. Forecast local ethnic populations in England, based on 2011 Census results, drift away from 2011-21 demographic events. To correct this, we adjusted the forecasts to ONS mid-year population estimates. The 2021 results for the raw and adjusted forecasts were tested against the post 2021 census mid-year populations. An additional step will be to adjust the forecasts to re-based mid-year populations, due for release by ONS shortly. Data. Published journal papers describe the official data sources for components, estimation methods and assumptions used in the forecasts. Methods. To forecast local authority ethnic populations, we used a bi-regional cohort-component model, designed to handle migration as an event. A look up matrix was used to harmonize ethnic populations to a 2021 local geography. Mid-year populations were converted by ONS to a 2021 geography. Findings. For England, the ethnic composition of the forecasts for 2021 and the 2021 Census were close, with a correlation of 0.99. The forecast total population at mid-year 2021 was 57.45 million, 1.6% higher than the 2021 Census. Differences resulted from failure of life expectancy to improve after 2015, Brexit from 2016 affecting international migration and the Covid-19 pandemic in 2020-21. Further work will use the re-based mid-year populations for adjustment and provide results for the 309 local areas used in the forecasts.

See extended abstract

 Presented in Session P3. Migration, Economics, Policies, History