Policy paper

Local Government Outcomes: Statistical Neighbours model

Published 9 February 2026

1. Local Outcomes Framework: Statistical Neighbours model

This document is being published alongside the launch of the Local Outcomes Framework.

To help comparisons between local areas and interpret data in the Framework, we will use a local authority statistical neighbours model. A statistical neighbours model identifies which local authorities are most similar to each other based on selected contextual characteristics. Using such models is common practice in Local Government to aid meaningful comparisons and benchmarking. 

This new MHCLG model works by comparing local authorities across a range of defined indicators, such as urbanicity/rurality and levels of deprivation. Local authorities with the most similar profiles across these indicators are identified as statistical neighbours.  

Creating a new model provides flexibility for the model to be tailored to the factors most relevant for the outcomes and metrics captured within the Framework. We are working closely with the Office for National Statistics’ (ONS) Local Statistics and Analysis Division, building on their nearest neighbours method and code to produce a model with underlying data and methods best suited for use with the Framework. As a result, the statistical neighbours used for the Framework will differ to those used in other areas, for example in the DfE’s Children’s social care dashboard. 

When the Framework outcomes and metrics are published on gov.uk on a digital tool, users will be able to compare the outcomes and metrics for a given local area to those of its statistical neighbours. This document provides detail on the statistical neighbours methodology and next steps in model development ahead of the tool launch. 

2. Methodology 

Determining Statistical Neighbours: 

Statistical neighbours are obtained by calculating how similar each local authority is to every other local authority across the range of contextual factors. Specifically:  

• Each local authority is assigned a profile consisting of its associated values across the chosen contextual indicators, creating a multi-dimensional snapshot of each authority. 

• Similar local authorities are then identified by calculating the Euclidean distance between these profiles, which is a mathematical measure of how similar the authorities are across all the contextual indicators simultaneously. 

• Authorities with the smallest Euclidean distances between them have the most similar profiles and are considered statistical neighbours. 

To ensure no single contextual indicator dominates the model, the indicators are pre-processed by capping outliers at the 1st and 99th percentile, and standardised such that each contributes equally to the calculation.  

Matching restrictions by local authority type are applied, to ensure that counties are matched with counties, non-metropolitan districts with non-metropolitan districts, and single-tier authorities with single-tier authorities. This ensures that neighbours are comparable in both context and service delivery. 

Selecting contextual indicators: 

• The indicators that are most relevant to the priority outcomes in the Framework were identified, starting with those used in the ONS model. 

• The indicators were tested for data quality and their technical validity assessed using principal component analysis and correlation testing in order to understand how they interact with other indicators. 

• The indicators were considered and iterated in collaboration with technical experts both within and external to MHCLG.  

• Judgement plays a part – care was taken to balance the number of indicators in the model and examine which Priority Outcomes they exert influence over – to avoid over-adjusting, whilst ensuring comparisons are meaningful. 

This approach ensures the proposed indicator set is both technically robust and grounded in relevant expertise. MHCLG will continue engagement with experts to refine and finalise the indicator list and invites views from the Local Government sector on the statistical neighbours model. 

3. Next steps 

The statistical neighbours model will be used in the Framework digital tool due to go live in 2026, following quality assurance and peer review.

Interested parties can provide feedback on the model and its use with the Framework in writing by email (localoutcomesframework@communities.gov.uk) by 31st July 2026. Responses will be reviewed and taken into consideration for any future iterations of the model to ensure the model represents and balances the different contexts within which councils operate to the best degree possible. 

Table: Draft Indicator List 

The proposed (draft) list of indicators is: 

	Indicator	Description	Theme	Source
1	Deprivation	English Indices of Deprivation (2025)	General contextual	MHCLG
2	Proportion of population identifying as white	Proportion of population identifying their ethnic group within the high-level “White” category	General contextual	ONS
3	Population Size	Mid-year population estimates	General contextual	ONS
4	Percentage of population in rural areas	Proportion of population living in rural output areas	General contextual	ONS
5	Employment Rate	Percentage of people aged 16 to 64 in employment	General contextual	ONS
6	Education rate	People aged 19 and above participating in further education (inc. apprenticeships), per 100,000 people	General contextual	DfE
7	Qualification rate	Percentage of population with Level 3 or above qualifications	General contextual	ONS
8	Digital connectivity	Percentage of premises with access to gigabit capable broadband coverage	General contextual	Ofcom
9	Gross value added per hour worked	Estimated value of goods and services produced per hour worked (£)	General contextual / Economic	ONS
10	Business births	New enterprises as a share of active enterprises	General contextual / Economic	ONS
11	High growth businesses	Percentage of enterprises classed as having “High growth” out of active enterprises with more than ten employees	General contextual / Economic	ONS
12	Proportion of population at working age	Proportion of working-age population (16 to 64) to non-working-age population	General contextual / Economic	ONS
13	Income	Gross median weekly pay (£)	General contextual / Economic	ONS
14	Proportion of population renting privately	Percentage of households classes as “Private rented”	Relevant to Housing, Multiple Deprivation,  Homelessness & Rough Sleeping	ONS
15	Percentage of children in relative poverty	Percentage aged under 16 living in relatively low-income families	Relevant to Keeping Children Safe, Best Start in Life,  Every Child Achieving and Thriving, Homelessness & Rough Sleeping, Health and Wellbeing	DWP
16	Percentage of pupils with Special Educational Needs (SEN) or Education and Healthcare Plan (EHCP) support	Percentage of pupils in receipt of SEN support or with an EHCP	Relevant to Keeping Children Safe, Best Start in Life,  Every Child Achieving and Thriving	DfE
17	Housing repossession rate	Private landlord possession claims per 100,000 households owned by a private landlord	Relevant to Housing, and Homelessness & Rough Sleeping	MoJ
18	Proportion of population without a disability	Percentage of population not disabled under the Equality Act 2021 (age-standardised proportion)	Relevant to Adult Social Care, Multiple Deprivation, Transport, Homelessness & Rough Sleeping	ONS
19	Housing affordability	Ratio of median house price to median gross annual workplace-based earnings	General contextual / Economic	ONS
20	Taxbase per household	Average council tax per dwelling (£)	General contextual / Economic	MHCLG
21	Area Cost Adjustment	Measure of local variation in the cost of delivering public services	General contextual / Economic	MHCLG