Open consultation

H: Technical Annex on the Fire and Rescue (Fire) Relative Needs Formula (RNF)

Updated 20 June 2025

Applies to England

1. Introduction

The Fire RNF was last updated prior to initial rollout in 2013/14 (hereafter the “current Fire RNF”). The government have been exploring updating the data used in this formula (hereafter the “updated Fire RNF”). This means keeping the same structure and formula components and, for the most part, keeping the data source and methodology used to generate scores for those components constant. Published information relating to the data and methodology used in the current Fire RNF can be found here. There are places where updated data from the same source was either unavailable or where continuing to use the data source would go against the principles of funding reform. In these instances, alternates were explored and tested with a technical reference group comprised of experts in funding formulae from around government. comprised of experts in funding formulae from around government.

2. Proposed formula structure

The proposed updated Fire RNF maintains the structure of the current Fire RNF. Firstly, a per capita need value is calculated for each Fire and Rescue Authority (FRA) using a basic amount and top-ups based approach as follows:

FRA need per capita = basic amount + (top-up1 * weighting1) + (top-up2 * weighting2) + (top-up3 * weighting3) + …

The resident population of the FRA is then applied to this per capita need figure to scale it up and give each FRA a RNF score:

FRA RNF score = FRA resident population * FRA need per capita

The Area Cost Adjustment (ACA) for the Fire RNF is then applied to give each FRA a final score:

FRA final score = FRA RNF score * FRA Fire ACA

The RNF share is then calculated for each FRA:

FRA RNF share = FRA final score / sum of all FRA final scores

The proposed updated Fire RNF maintains the use of the 7 top-ups in the current Fire RNF. What they are, reasoning for their inclusion and their average relative contribution to FRA RNF scores are shown in Table 1 below. The average relative contribution of each top-up to the formula has been generated by calculating the mean weighted value in the current RNF for each top-up, summing them with the basic amount and multiplying that by the mean population to generate an average FRA RNF score. Each top-up mean value is then divided by the average FRA RNF score to calculate what proportion of the score it is responsible for.

Table 1. Name, reason for inclusion and average relative contribution of each of the 7 top-ups contained within the current Fire RNF.

Top-up	Reasoning	Average relative contribution
Basic amount	Used to ensure a baseline amount per capita is allocated to each FRA	58%
Coastline	FRAs with extensive coastlines have decreased ability to call on neighbouring services for support in the event of a major incident.	3%
Density	More people in less space can lead to an increase in fire risk.	12%
Sparsity	Travelling across large, remote areas to attend incidents increases pressure (and cost) on services.	2%
Deprivation	There is a large body of evidence to suggest a relationship between increased deprivation and increased fire risk.	10%
High risk	More Control of Major Accidents and Hazards (COMAH) sites present an increased risk of a major incident.	2%
Property and societal risk	Certain property types have a greater frequency of fires and specifically fires which result in rescues (societal risk) and/or damage (property risk), so FRAs with more of those properties are subject to increased risk.	7%
Community fire safety	Some areas or populations are associated with a greater need for education around fire safety which incurs a cost to the FRA responsible.	6%

3. Changes from the current Fire RNF

Although care was taken to derive updated top-up values from the same data sources used in the current Fire RNF, this was not possible for all top-ups (for example, due to changes in the benefits system effecting use of benefit claimant data for the deprivation top-up). The indicators used in the proposed updated Fire RNF are discussed below.

Resident population

The proposed updated Fire RNF will calculate projected residential population using the mid-2023 population estimates for each FRA aggregated from Local Authority District (LAD) level Office for National Statistics (ONS) figures for the total usual residents in households and communal establishments population.

Should ONS release 2022-based subnational population projections prior to the rollout of new allocations, we are considering updating the Fire RNF to include these accordingly, including updating any top-up values that are dependent on the measure of residential population.

Coastline top-up

Coastline top-up values will be calculated as the FRAs length of coastline per head of population. Data on length of coastline (to the nearest 1000 metres) is at low water using information on 1991 administrative areas from the Boundary Line Product provided by Ordnance Survey (OS). This is the exact same data as used in the current Fire RNF. Where FRAs have merged since this data was used in the current Fire RNF (for example, Hampshire and the Isle of White), the coastline lengths for the two constituent FRAs have been summed to give a coastline length for the merged FRA. More up-to-date options for data sources were explored but issues arose due to the inclusion of tidal estuaries in more recent data. Whilst there could be an argument that tidal estuaries also present additional risk to FRAs, this was not a top-up rationale included in the development of the current Fire RNF and this update only seeks to update data in the formula, not the structure, components or rationale. The underlying assumption behind using the original data source is therefore that coastlines have not significantly shortened or lengthened differentially across FRAs since 1991 (i.e. some FRAs have not seen large coastline length increases whilst others have seen large decreases, not because of boundary changes). Total estimated resident population will be as calculated above.

Density top-up

Density top-up values will be calculated in the same way as in the current Fire RNF, as the sum of the estimated number of residents per hectare for each Census 2021 Output Area (OA) within the FRA, projected forward to 2023 using all-England population change rates between 2021 and 2023 changes in mid-year estimates for those years), multiplied by that OA’s share of the usually resident population in the area. This result is then divided by 10 as in the current Fire RNF methodology.

Sparsity top-up

Sparsity top-up values will be calculated using the same population data as density above, converting the figures from square kilometres to hectares and then using the same methodology and judgement-based weights as in the current Fire RNF. As such values are calculated as the sum of (i) and (ii):

i) Super sparsity – as per the formula below where all population estimates are as defined above:
3 multiplied by the resident population of those OAs with less than or equal to 0.5 residents per hectare, divided by the total resident population of the FRA

ii) Sparsity – as per the formula below where all population estimates are defined above.
The resident population of those OAs with more than 0.5 but less than or equal to 4 residents per hectare, divided by the total resident population of the FRA

Deprivation top-up

The current Fire RNF calculates deprivation top-up values using a Risk Index which combines data on: the labour force, benefit claimants, apprenticeship enrolments, New Deal programme participants and mortality. The proposed updated Fire RNF will instead use the 2021 Census data on household deprivation (table TS011). This is because the New Deal programme no longer exists and has not been replaced by an equivalent, apprenticeship pay was brought into line with minimum wage for 18-year-olds in 2022 (previously it was lower), and the introduction of Universal Credit as part of benefits reforms means it would not be possible to replicate the specific benefits included in the Risk Index. Redevelopment of the index was explored but, in line with the reform principles of transparency and simplicity, the decision was made to switch to using Census data instead.

The Census household deprivation measure captures 4 dimensions of deprivation (employment, education, health and disability, and housing), scoring households on a binary (yes/no) for each they meet the criteria for to be considered deprived. The deprivation top-up value will be calculated as the number of households within the FRA which meet any 3 or all 4 dimensions of deprivation, as a proportion of the total number of households within the FRA.

High risk top-up

High risk top-up values will continue to be calculated as the number of top tier Control of Major Accident Hazard (COMAH) sites within the FRAs boundary per head of population. This will be as at April 2024 and derived from data supplied directly to the Department by the Health and Safety Executive. Total estimated resident population will be as calculated above.

Property and societal risk top-up

The current Fire RNF calculates property and societal risk top-up values as the sum of the number of buildings by building type within an FRA multiplied by (i) property risk frequency by building type and (ii) societal risk frequency by building type, per head of population. For this, data on building volumes (by type) from the Valuation Office Agency and risk frequencies from the Fire Services Emergency Cover Toolkit are used. That latter is an IT-based tool designed to enable FRAs to assess the risks from fire and other incidents. It has not been updated since 2012 and is no longer in regular use. There is no single, up-to-date equivalent as services have developed their own, bespoke risk modelling strategies.

In the proposed updated Fire RNF the calculation will remain the same (with the addition of weightings to the risk frequencies as detailed below) but data sources will change. Total estimated resident population will be as calculated above. Building volumes will be from the Ordinance Survey (OS) Points of Interest dataset (supplied directly to the Department by OS). Moving to use OS data will bring the Fire RNF closer in line with the methodology used to allocate the Protection Uplift Grant. It also maximises existing data flows into the Department from OS and opens the possibility for further exploration of risk as OS data develops. Risk frequencies will be generated using Incident Recording System data held and published by the Department.

Property risk frequency will be the likelihood a fire in each building type would result in damage to property. It will be calculated as the average estimated extent of fire damage by building type during 2021/22, 2022/23 and 2023/24. Societal risk frequency will be the likelihood a fire in each building type would result in the need for the responding FRA to engage in a rescue. It will be calculated as the average number of rescues by building type during the same period.

Risk frequencies will be weighted before being applied to building volumes. This is necessary to ensure 1 square metre of property damage is not weighted the same as 1 rescue. In the current Fire RNF the average property risk frequency is 11% larger than the average societal risk frequency. The weightings used adjust using this by: (i) calculating the new average societal risk frequency, (ii) multiplying this by 1.11 and considering the result the ‘target’ average property risk frequency, (iii) calculating the reduction necessary to the real average property risk frequency to bring it to the ‘target’ one – this is the weighting to be applied.

Community fire safety top-up

The current Fire RNF calculates community fire safety top-up values using a bespoke index which combines data on: primary school aged learners (and their attainment), population of people aged 65+ and the number residents identified as living in areas with a greater need for fire safety education. The latter is sourced from Acorn data supplied by CACI Ltd. This is a commercially available geodemographic segmentation tool which categorises every UK postcode area into one of 65 Acorn types to describe their demographic make-up. The Acorn types selected for inclusion in the current Fire RNF broadly cover postcodes which contain higher proportions of children, students, the elderly, ethnic minorities and low-income families.

In line with the reform principles of transparency and simplicity, the decision was taken to move to using 2021 Census data on household composition (table TS003) for the proposed updated Fire RNF instead of this index. Community fire safety top-up values in the proposed updated Fire RNF therefore will be calculated as the total number of households within an FRA that contain students, dependent children or people aged over 65, as a proportion of the total number of households within the FRA.

Top-up weightings

All top-up values are weighted before they are summed with the basic amount and then multiplied by population. For the updated Fire RNF, the relative contribution each top-up makes to the overall formula has been kept the same as in the current Fire RNF so appropriately adjusted versions of the weightings used in the current Fire RNF were used again. Adjusted weightings were created by calculating the mean raw score for each top up in the current Fire RNF and in the updated Fire RNF, calculating the change between the two scores and applying the reverse to the top-up. This method is only adjusting for issues arising from using different data sources and methodologies for some of the top-ups.

Area Cost Adjustment

The government is proposing to apply an updated ACA for the Fire and Rescue formula in line with the approach taken across the other RNFs. However, due to the structure of the proposed updated Fire RNF being based on the current one, this presents a few key issues not found in other formulae.

Some of the top-up weightings of the current RNF are based on modelling which incorporated the current ACA. As the proposed updated RNF will maintain the current RNF top-up weightings, this means the final RNF shares of each FRA will be determined partially by the ACA used in 2013-14, and partially by the updated ACA.

In addition, the updated ACA accounts for journey times as a component of labour cost through its Accessibility adjustment and through the Remoteness adjustment. The current Fire RNF (and therefore the proposed updated one too) captures metrics which are strongly correlated with these differences in travel times through population density and sparsity. As such, the impact of journey times on overall FRA formula shares will be more influential and potentially double counted to some degree. The government is currently exploring whether it would be possible to mitigate some of the issues caused by the interaction between the updated ACA and the Fire RNF.

4. Calculating proposed Fire RNF shares

We propose to calculate the shares of the Fire RNF for each FRA as follows, using data on top-ups discussed in the previous section:

i. Basic amount of 2.7837, plus
ii. Fire Coastline top-up of 1.2861 multiplied by coastline length per head of population; plus
iii. Fire Density top-up of 0.1258 multiplied by population density; plus
iv. Fire Sparsity top-up of 0.1245 multiplied by population sparsity; plus
v. Fire Deprivation top-up of 14.1194 multiplied by proportion of households meeting 3 or more dimensions of deprivation; plus
vi. Fire High Risk top-up of 20,257.1126 multiplied by the number of COMAH sites per head of population in an FRA; plus
vii. Fire Property and Societal Risk top-up of 251.3872 multiplied by the Property and Societal Risk score; plus
viii. Fire Community Safety top-up of 0.5063 multiplied by the Community Safety score.

The above result is then multiplied by:

i. FRA residential population, and
ii. ACA for the Fire and Rescue formula.

In order to calculate the final RNF share of an FRA, its results from the above should be divided by the sum of these results for all FRAs.