Research and analysis

Economic and social cost of fire

Published 29 June 2023

Applies to England

Foreword

I am delighted to introduce this report on the economic and social cost of fire in England. Since fire and rescue policy moved to the Home Office in 2016, the department has been keen to update the government estimates, which were last published over a decade ago. This report is the culmination of that work and represents a substantial addition to the fire economics evidence base.

I hope that this report will be widely used by fire and rescue services, fire industry professionals, and other readers, to make better evidence-based decisions. The report is a starting point, along with other research reports published last year on evacuation and fire related fatalities, of a growing and vitally important fire research function in central government that I hope will continue to deliver in outputs and insights that can drive value for the taxpayer. It is my aim for the Home Office to take a leading role in developing the fire evidence base, which with our support will continue to grow and be vital in informing policy and operation decisions that help improve public safety, save lives, and protect against the damage that is caused by fires.

Simon Palmer

Chief Analyst and HOAI director

Acknowledgements

This report was authored by Jack Pickering (senior responsible analyst), Will Beall and Will Phillips, with support from Alice Plumridge, Pippa Cousins and Phoebe Maguire Hamblett.

The authors would like to give special thanks to all who attended our internal working group and external advisory group. This included individuals from fire and rescue services, the National Fire Chiefs Council, other government organisations and industry experts whose support was invaluable.

We would also like to thank all the individuals who supported with the analytical quality assurance of this work, and the individuals who peer reviewed the final report.

For any questions relating to this report, please email: FireEconomics@homeoffice.gov.uk.

Executive summary

This report estimates the total annual economic and social cost of fire in England using data on fires that occurred in the year ending March 2020. This significant addition to the fire economics evidence base provides an up-to-date robust and reliable estimate, which is vital in understanding the impact of fires in England. The purpose of this report is to inform policy development, operational decision-making and be a point of reference for analysts and individuals working with, and for, fire and rescue services (FRSs).

The estimated total economic and social cost of fire in England, in year ending March 2020, was £12.0 billion, of which £3.2 billion is marginal cost. Marginal cost is defined as the cost incurred following a fire, that would be directly impacted by the change in the number of fires. The marginal unit cost can be defined as the change in total economic and social cost from one additional fire occurring. Therefore, marginal cost is seen as a better cost figure for estimating the value and economic benefit in reducing the number of fires that occur. Total cost also includes anticipation spend, which is more related to fire risk as opposed to fire incident numbers. The costs are split into 3 overarching areas:

Anticipation – measures designed to either prevent fires from occurring or protective measures to mitigate the damage and impact of fires.

Consequence – direct and indirect costs that occur as a result of fire, such as property damage, loss of business, human injury, and fatalities.

Response – cost of fire and rescue services responding to incidents.

Table M1: Total economic and social costs of fire, year ending March 2020

Cost area	Total cost	Average Unit cost (all fires attended)	Total cost (primary fires)	Unit cost (primary fires)
Anticipation	£8.8bn	£57,100	-	-
Consequence	£3.2bn	£20,500	£3.1bn	£45,000
Response	£74m	£500	£59m	£900
Total	£12.0bn	£78,000	-	-
Marginal cost	£3.2bn	£20,900	£3.2bn	£45,900

Table M1 notes:

1. Total costs are the sum of anticipation, consequence, and response. Marginal costs are the sum of consequence and response. Unit costs rounded to nearest £100.
2. Anticipation costs are not split out by fire type, and are not calculated for primary fires.

Table M1 shows a breakdown of the total cost by these 3 areas. The largest costs of fire are in anticipation, with approximately:

• £4.6 billion coming from defensive expenditure in buildings
• £2.0 billion coming from defensive expenditure in consumer goods
• £1.4 billion coming from fire and rescue service expenditure in anticipation and preparation for fires

Other major cost areas come as consequences of fires, with approximately:

• £2.0 billion coming from property damage
• £0.4 billion coming from physical and emotional harms
• £0.3 billion coming from road vehicle fire damage

Unit costs are also presented, which are calculated as the average cost per fire attended by FRSs. The average cost per fire attended (total cost from anticipation, consequence, and response, divided by the total number of fires attended) is £78,000. When removing anticipation costs, the unit marginal cost of all fires attended is £20,900, and the cost of a primary fire attended is £45,900. These costs are split by fire type in table M2, with more detail on the costs in the main body of this report.

This report lays out the total cost estimates in the introduction and main results sections, and then explains in detail the calculations behind the total costs reached. This level of detail has been provided for full transparency and to ensure the report is of as much use as possible to readers, who may want to better understand the approach and certain elements of the total cost. This report tries to monetise as many of the impacts of fire as possible using economic theory and appraisal techniques. However, it is not possible to monetise all the social impacts of fire. Where impacts of fire cannot be monetised, they are referenced in section 6: Future developments.

This report is the best available current estimate for the total economic and social cost of fire and is the culmination of a significant amount of analysis. It builds on and improves the seminal work on this topic published by the Department for Communities and Local Government (DCLG, now DLUHC) in 2011 (DCLG, 2011a, DCLG 2011b) and follows a similar overall methodology to other Home Office published appraisal “cost of” reports, such as those published on crime (HO, 2018) and domestic abuse (HO, 2019). However, this report uses novel analytical approaches and methodologies, and attempts to cost areas which have not been included in previous reports on the cost of fire. Accordingly, it is not appropriate to directly compare total figures. Although this report is the most robust and accurate estimate possible, some significant uncertainties exist (especially in the anticipation and property damage figures) and assumptions have had to be made. These could be improved on with further specific work and are referenced throughout the report where relevant, as well as in the ‘Future developments’ chapter.

Table M2: Total direct marginal costs of primary fires by fire type, year ending March 2020

Fire type	Total cost	Unit cost (primary fires)
Other Buildings	£1,779m	£124,200
Dwellings	£922m	£32,400
Other Outdoors	£90m	£16,700
Road Vehicles	£364m	£17,700
Total (Primary)	£3,155m	£45,900
Deliberate	£927m	£48,400
Accidental	£2,228m	£44,900

Table M2 notes:

1. All total costs rounded to the nearest £1 million, and all unit costs rounded to nearest £100.

1. Introduction

The aim of this report is to calculate the total annual economic and social cost of fires in England using data on fires that occurred in year ending 31 March 2020. The purpose of the report is to improve our understanding of the economic impact fires have on individuals, businesses, and society, which is currently a significant evidence gap.

This report represents the first comprehensive attempt by the Home Office to estimate the economic and social cost of fire in England and to our knowledge, the only full assessment of this cost since DCLG published an estimate in 2011 (DCLG, 2011a, DCLG, 2011b).

This report is a significant update on previous estimates, with overarching methodological changes made throughout, and numerous costs (such as environmental impacts) monetised for the first time in published economic analysis on fires. It should not be compared to previous publications on this topic due to large methodological differences.

The report is a significant addition to the fire economics evidence base and aims to be a useful tool for government decision makers and analysts, FRSs, and wider stakeholders working in the fire sector. The figures presented can be used in economic appraisal, the evaluation of policy and operational activity (including prevention and protection activity), and financial assessments. Ultimately, this report aims to provide policy and operational decision makers with an improved understanding of the cost and impact of fire, which will assist their efforts to reduce the number of fires and the harm that they can cause society.

1.1 What is included?

The analysis breaks down the cost of fire, by fire type and whether it is classed in the incident reporting system (IRS) as accidental or deliberate. It is beyond the scope of the report to split the cost of fire into more detailed location categories and causes.

The report splits the cost of fire into 3 main cost areas, each of which contain multiple separate cost categories. This approach aligns with other Home Office “cost of” reports, and previous global research into the cost of fire. The report aims to monetise as many impacts of fire as possible and includes both economic and social cost. However, it has not been possible to monetise all the impacts and costs of fire due to evidence and data gaps. These are highlighted as non-monetised costs, and a full explanation of the chosen scope is in section 1.2. In addition, the report takes a proportional approach to deciding which costs should be monetised and which should be considered out of scope. Only impacts that are directly, or in some specific instances indirectly, impacted by fires are included to prevent incorrectly overestimating the cost of fire. The main costed areas are explained below.

Anticipation

Defensive expenditure in buildings

Defensive expenditure in buildings includes any cost incurred from measures to improve fire safety and protect against fires in buildings (such as fire doors and compartmentation structures). It includes the cost of incorporating active and passive fire protection in buildings, and from designing buildings in a fire safe way.

Defensive expenditure in consumer goods

Companies incur costs from making sure that consumer goods they sell (such as electrical appliances, furniture, and motor vehicles) meet fire safety standards and comply with the appropriate legislation to ensure they are fire safe. This may involve the cost of fire safety testing, adding protective markings, applying fire retardants, or using more expensive, less flammable or combustible materials.

Carbon emissions from fixed fire suppression systems

This is the cost in terms of carbon emissions from fixed fire suppression systems in some commercial buildings that aim to prevent fires that may have required FRS attendance.

Insurance administration

The expenses incurred from administration of developing insurance policies, and the running costs of insurers to facilitate the delivery of these policies, is considered to be an economic cost to society because these resources could have otherwise been used in alternative productive activities.

Fire training and awareness in workplaces

There is an economic cost of time spent from individuals undertaking workplace fire safety activity.

Expenditure by fire and rescue services

The proportion of the total FRS budget that is estimated to be attributed to activity in anticipation of having to deal with fires. This includes training, resilience requirements, and any activity that aims to prevent and protect against fires occurring.

Consequence

Physical and emotional harms

This is the reduction in the quality of life of those who are victims of fire. It includes direct physical and emotional harms from injuries, fatalities, and potential emotional harm suffered by those who are rescued from fires.

Property damage

This is the cost of damage to properties as a result of, fire, water, and smoke damage following a fire incident. It includes direct damage to the property itself from the incident, and also wider associated damages, such as to contents, stock, machinery, and business interruption.

Road vehicle damage

This is the cost of damage to road vehicles as a result of fire.

Lost output

The cost of lost output as a result of fires is split into 2 broad categories. The first is lost output following property damage in businesses, which could stop individuals from being able to work and lead to unemployment. The second category comes from victims of fire, which causes lost productivity when individuals take time off work and/or are unable to work, and potentially reduced productivity whilst at work following injuries.

Health services costs

This is the cost to health services in dealing with the physical and emotional harms of fire. These are mainly based on NHS unit costs of treatment, and include ambulance costs, medical procedure costs associated with physical harms, and counselling costs associated with emotional harms.

Environmental costs

This includes the cost to society of carbon emissions from fires, the reduction in air quality from fires, and also the cost of environmental damage from outdoor fires, proxied using the cost of repairing and restoring damaged land.

Insurance administration

The expenses incurred from the administration and management of claims is considered to be an economic cost to society because these resources could have otherwise been used in alternative productive activities. This falls in consequence because these costs are only incurred once a fire occurs, and an insurance claim is made.

Crime specific costs

Some deliberate fires will be criminal offences and occur as a result of criminal behaviour. The crime specific costs of these arson offences are included in this report and are based on uprating the estimates and methodology in the Home Office cost of crime publication (HO, 2018). There is a cost to the police and criminal justice system (CJS) of dealing with arson crimes, and additional costs associated with providing support to victims of arson.

Response

Fire and rescue service response costs

This is the direct cost of FRSs responding to fire incidents, which is considered to be an economic cost (opportunity cost) to society because in the absence of fire incidents, FRS resources could be used in other productive activities. It includes the time spent by firefighters attending incidents, non-employee costs directly incurred in response (such as vehicle wear and tear and fuel), and control costs. It is not equal to the full cost of the FRS and does not include fire related expenditure and time spent by FRSs that are not associated with responding to incidents but are required to be able to respond to incidents (resilience). These costs are included in the anticipation costs section of the report. Costs related to non-fire incident response, preparedness and fire false alarms are excluded from the total cost of fire estimate but are calculated in this report.

Additional cost areas, and non-monetised impacts

A wide range of costs are included in this report, and we consider this to be the most comprehensive estimation of the economic and social costs of fire in England available. Inevitably, not all economic and social costs could be monetised in this report. There are some impacts where there is insufficient data and research for them to be accurately monetised. These “non-monetised” costs are identified throughout the report, and further outlined in section 6. Further research could look at costing these impacts.

1.2 Scope of the report

It is important at the outset of this report to define the scope, and what the report aims to include, and not include. The main scope of the report is:

The financial year to March 2020 is used as the base year for all the calculations in this report, and all costs are in year ending March 2020 price years

Year ending March 2020 was chosen as the base year as it is the most recent year with readily available published data that is not impacted by the coronavirus (COVID-19) pandemic in England, and subsequent lockdowns^{[footnote 1]}. This report is seminal work that represents a starting point for this sort of analysis, and the calculations could be updated in future years to provide annual total estimates for year ending March 2021, once data is available.

Costs incurred in year ending March 2020 that aim to prevent or protect against fires are included as anticipation costs, with consequence and response costs being based on fires that occurred in year ending March 2020

The total costs calculated are an estimate for the annual cost of fire in England. When considering anticipation costs, only expenditure incurred in year ending March 2020 is included. Spend incurred in previous years that still aims to prevent or protect against fires in year ending March 2020 (for example, the installation of a smoke alarm system in year ending March 2019 that is still in place in year ending March 2020) is excluded, and it is noted that the benefits of fire prevention activity that occurred year ending March 2020 may continue in future years, potentially impacting the number of fires in future years. This nuance is important to be aware of when considering anticipation costs over time, and how spend on anticipation relates to consequence and response costs incurred. Consequence and response costs are based on fires that occurred in year ending March 2020. Costs incurred from these fires beyond year ending March 2020 (for example, from fatalities or longer-term lost output), are included, and discounted back to year ending March 2020 prices appropriately.

Only fires attended by FRSs are included

This ensures that the unit costs calculated can be directly compared to Home Office published incident recording system (IRS) data which includes all fires attended. Small fire incidents that FRSs are not aware of and/or do not require an FRS response are excluded because these are expected to have no cost, so would skew the unit cost estimates downwards if included. FRSs respond to all fires they are made aware of, and we assume that FRSs are notified of all fires that cannot be extinguished by individuals or do not go out by themselves. This assumption is supported by the English Housing Survey (DCLG, 2015) which found that in year ending March 2014, 262,000 English households experienced one or more fires inside their home or on their property, compared to the 31,912 dwelling fires attended by FRSs in England (HO, 2023, FIRE0102). However, the majority of these fires experienced by households (71%) went out themselves, or were put out by individuals, not by the FRS. Only 7% led to any fire related injuries. It is therefore assumed that all injuries caused by fires come from fires attended by FRSs, and the impact of fires not attended is generally minimal.

The report includes total and unit cost estimates

These estimates are split into certain different fire types and causes (deliberate/accidental) where possible. Anticipation costs are not split out into fire type due to a lack of data. Unit costs are presented both in terms of all fires attended, and all primary fires attended for completeness and to account for differing severities. Secondary fires make up a large proportion of overall fires attended but are generally lower impact and have limited or no consequence costs, so unit cost estimates are presented both with, and without, these fire types included.

The report only considers the cost of fires and does not aim to cost FRSs as a whole, or the cost and impact of all FRS activity

FRSs undertake activity that is not fire related. In year ending March 2020, FRSs attended 172,207 non-fire incidents (HO, 2023, FIRE0102), including approximately 31,000 road traffic collisions and 16,000 flooding incidents (HO, 2023, FIRE0902). However, the cost of these incidents is beyond the scope of this report. The report does consider the cost of FRSs responding to false fire alarms in the response chapter. However, these costs are not included in the overall headline cost of fire.

The report only looks at the economic and social cost of fire and does not seek to calculate the economic value of fire and rescue activity

Although one of the uses of the report may be in assessing the impact of FRS activity, especially related to prevention and protection, this report does not aim to quantify the large value that FRSs provide to society. Other reports do look to assess this (NFCC, 2022), but this report strictly considers the costs of fires, and not the benefits of FRSs.

The cost estimates in this report use the best available evidence and data at the time of publication to assess the cost of fire in year ending March 2020. The total and unit cost estimates are inevitably sensitive to changes in fire trends, organisational developments, and technological changes. These factors should be considered when using the analysis and are especially pertinent for future research that may look to update the cost estimates. The potential specific impacts of COVID-19 and lockdowns (year ending March 2021), rising inflation (CPI reached 11.1 per cent in the year ending October 2022 (ONS, 2023)) and costs incurred from additional cladding remediation and the Fire Safety (England) Regulations 2022 are not considered in the costs. The impact of inflation, and recent supply issues in some sectors, are especially important to note for the construction costs included (which in some cases have seen marked rises), and the property damage costs in this report.

The costs in this report are the most robust and accurate estimates possible. However, some assumptions have had to be made, and some significant uncertainties exist in the figures. The analysis could be improved with further specific work to strengthen the evidence based used in some sections. Uncertainties in the evidence used and assumptions applied are identified throughout the report where relevant. The final chapter sets out comprehensive recommendations for how the work could be improved if further, specific analysis was undertaken.

1.3 Interpreting and using the report

The figures in this report will help policymakers and practitioners better understand the full impact of fires on individuals and society. They can also be used to assess the benefits of reducing the number of fires and assessing the cost-benefit of interventions. With multiple potential users, and multiple unit costs presented in this report, user guidance is provided to ensure that the costs are correctly used. The unit costs are based on fires (all fires, or just primary fires) attended by FRSs, so can be used in conjunction with statistics published by the Home Office.

1.3.1 Total and unit costs

Costs throughout this report are presented either as total costs (overall cost of this type of impact on society) or as unit costs (the cost per fire). Total costs should be used when estimating the total impact to all individuals, or across society, whereas unit cost should be used when assessing the impact of a single fire. The latter can be used in policy appraisal and the evaluation of interventions.

1.3.2 Marginal unit costs

Costs in this report are split into overall totals and marginal costs. The difference between the 2 is that marginal costs do not include anticipation costs. Total unit costs could be used for illustrative purposes when looking at the cost of fire in England at a single point in time (year ending March 2020), and to provide context to total cost numbers by showing them per fire. However, marginal unit costs are the most appropriate figures to use when assessing the cost, or benefit, of changing numbers of fire incidents attended by FRSs. For example, in policy appraisal, marginal unit costs should be used because these costs are directly impacted by fire numbers.

Anticipation costs are part of the overall cost of fire to society because the existence of fires means that expenditure is required to mitigate risk. However, these costs are a snapshot in time and should not be used in economic appraisal because they are not marginal costs.

Marginal cost in economics is defined as the change in total cost as a result of one additional unit being produced, and is often used when assessing production. For the purposes of economic appraisal, marginal cost is used to assess the impact of a policy, which, in this report’s context, may reduce or increase the number of fires.

Anticipation costs are not marginal because a change in the number of fires may not directly impact anticipation costs in a calculable way. The anticipation costs in this report are generally directly impacted by legislation, industry standards, or the risk posed by fires as opposed to the number of fires. The relationship is complex.

Using an extreme example, if there were no fires in a given year, consequence and response costs would be zero because they fall directly with fire numbers. It is much harder to estimate the impact on anticipation costs, although it is logical that these would not fall to zero in the same year or in a calculable timeframe. The main drivers of anticipation costs which are legislative and regulatory requirements on building and consumer goods safety, FRS protection and prevention activity and behavioural, climate and technological changes, are not driven by fire numbers, and would not change in that year. While anticipatory spending may decrease in the long-term with a decrease in the number of fires, in the short-term, it is very likely that changes in fire incident numbers, would not be met with equivalent changes in anticipation cost.

The relationship is further complicated by the potential relationship between anticipation costs and both consequence and response costs. We may expect that as more anticipation costs are incurred, the better fire prevention and protection is, and therefore fewer (or less impactful) fires occur. This would reduce consequence and response costs. Alternatively, as anticipation costs fall, the risk of fire impact increases, which could increase consequence and response costs. This further shows that anticipation costs shouldn’t be used in marginal cost calculations.

Significant, or long-term, changes in the number of fires could impact anticipation costs. In these circumstances it could be appropriate to account for anticipation costs in appraisal, however, in the majority of cases they shouldn’t be. Individuals using this report for appraisal should consider on a case-by-case basis whether anticipation costs should be included, with a clear rationale provided if included.

1.3.3 Different unit costs

Users of this report are advised to use the marginal or total cost that is most related to the incident type they are considering in their research question, rather than the overall unit cost of all fires. Primary and secondary, and dwelling, outdoor, and other building fires have very different unit costs, so users should consider which incident type their policy intervention will impact before deciding which marginal cost to use, and should be as specific as possible. Overall costs across all fire types should only be used if a more specific unit cost type related to the question cannot be found or is not appropriate.

1.3.4 Deliberate fires and arson

Costs in this report are split into deliberate and accidental fires using IRS data on the motive for a fire. Accidental fires are those where the motive for the fire was presumed to be accidental or unknown, whereas deliberate fires are those where the motive was ‘thought to be’ or ‘suspected to be’ deliberate and includes damage to one’s own or other’s property (HO, 2022b).

Arson offences are a subset of deliberate fires, so all arson offences are deliberate fires, but not all deliberate fires are arson. An arson offence is defined in the Criminal Damage Act of 1971 as ‘an act of attempting to destroy or damage property, and/or in doing so, to endanger life’. Arson offences are included within deliberate fires.

It is difficult to compare the number of deliberate fires and arson offences because they are recorded by separate organisations (FRSs and the Police) and are defined differently. This means there can be large differences in trends within and across areas. There are considerably more deliberate fires recorded by FRSs than arson offences, and not all arson offences lead to an FRS response (the police may be made aware of an arson offence after the fire has been put out, so it is not attended by FRSs).

So, the deliberate cost estimates in this report are not equivalent to the direct cost of an arson offence. They differ too much in recording and incidence to equate directly. However, there are some similarities. The crime specific costs in this report are based on a total cost of arson, so could be converted to an arson unit cost with appropriate incident numbers, and other costs (response, property damage and to some extent physical and emotional harms) will likely have a broadly equivalent unit cost between arson and deliberate fires, presuming that the arson offence is attended by FRSs. Further work is required to calculate a unit cost of arson that effectively accounts for recording differences between FRSs, the Police, and the crime survey, and any use of deliberate fires as a proxy for arson should only be made with sufficient caveats.

1.3.5 Adjusting the costs to future price years

The costs in this report are all in year ending March 2020 prices, however users may want to use these costs in economic appraisal, or policy analysis, to assess the impact of fire numbers changing in different price years. The Quality Adjusted Life Year (QALY) value of £70,000 should be held constant when adjusting price years (HMT, 2022 and DHSC guidance) so, to avoid overestimating the cost of fire, guidance and a tool on how to uprate if required is presented in annex table A1.

1.3.6 Additional guidance

The use of unit costs is standard across government economic appraisal, however, unit costs are less useful when there is a large range in actual observed costs. Although IRS data splits fires into incident types, there is a large range in the average cost and impact, even within incident types. Two fires will both be classed as dwelling fires in the IRS, even if one only resulted in a small amount of damage (low cost) and the other led to multiple fatalities and a large amount of damage (high cost). When assessing the cost of these two events, the unit cost, as presented in this report, has some use, although the value will be far from the true cost of both these fires. This report uses a large number of fires to reach an accurate unit cost, however, users must remember that all fires are not the same so some caution should be exercised if applying the unit cost to real life events.

Individual fires of interest should still be assessed separately by policymakers and practitioners, as some large fires with specific, large, or rare impacts could lead to costs of a magnitude much larger (or smaller) than the unit costs in this report. This report can be used as a framework for how to assess the cost and impact of a single fire, with some simple adjustments to the assumptions used made in individual use cases.

In addition, this report uses novel methodologies and monetises cost areas that previous reports by DCLG have not, making the two reports incomparable. So, direct comparisons between the total cost estimates in the two reports should be avoided. The total cost estimate in this report exceeds the estimate in the DCLG report, but this is not evidence of an increasing cost of fire in the period since that report. Instead, it is reflective of an improved, more robust, methodology, and improved ability to monetise impacts.

1.4 Data quality

The sources of data and assumptions used in this report are referenced and footnoted throughout. The majority of the data used in the consequence and response sections of this report come from the Home Office’s online Incident Recording System (IRS). This system allows FRSs to complete an incident form for every incident attended, be it a fire, a false alarm or a non-fire incident (also known as a Special Service incident). The online IRS was introduced in April 2009. Previously, paper forms were submitted by FRSs and an element of sampling was involved in the data compilation process. The online IRS makes some automatic checks, including completed data or time fields, but not all the data can be checked to ensure it is logically correct. FRSs have been provided with a priority list of fields that should be checked, before updating incidents from a status of ‘recorded’ to ‘published’. Only once FRSs have submitted an incident as ‘Published’ will the Home Office quality assure the data.

Data received by the Home Office undergoes a quality assurance process to ensure it is fit-for-purpose. The Home Office completes a ‘monthly monitoring’ process to examine the data for gaps and variance checks – any large differences or other data quality issues are identified and subsequently resolved with FRSs. The Home Office publishes a Methodology and Quality report on Fire and Rescue Incident Statistics, where more detail on the data collection and quality assurance process is available (HO, 2022c).

The data collected in the IRS is published by the Home Office on the GOV.UK website at regular intervals. As the IRS is a continually updated database, with FRSs adding incidents daily, a snapshot of the IRS was used for this analysis. This means that the statistics may not match all those held locally by FRSs, and future reports that aim to replicate this work may slightly revise the figures. This is particularly the case for statistics with relatively small numbers, such as fire-related fatalities. For instance, this can occur because coroner’s reports may mean the initial view taken by the FRS will need to be revised; this can take many months, even years, to do so. The incidents included in this analysis are those which were responded to by English FRSs (for response costs) and occurred in English FRS territory (for consequence costs), which means some data totals may differ slightly to published Home Office releases. In the vast majority of cases, an English FRS both responds to incidents in English territories, however there are a handful of incidents with Welsh or Scottish FRS responses.

On a quarterly basis, a snapshot of the IRS is taken, with additional quality assurance checks completed. This snapshot enables the replication of the statistics produced, and this provides additional historical scrutiny and facilitates further investigative analysis if required. The figures in this report refer to records of incidents that occurred up to and including 31 March 2020. This includes incident records that were submitted to the IRS by 17 June 2021, when a snapshot of the database was taken.

Many of the fields in the IRS, including those on fire spread and severity, which are used throughout this report, are completed by firefighters at the scene using their best estimates for the impact of the incident. This means the data has some inherent uncertainty in it, which must be considered when using the results of this report. However, it remains the best available source for fire and rescue incident data.

2. Main results

This chapter discusses the overall economic and social costs of fire in England. The costs are split into anticipation, consequence, and response, with total overall and unit costs presented at the end of the chapter.

2.1 Anticipation

Anticipation costs are defined as expenditure incurred in year ending March 2020 on measures to prevent or protect against fires. Total anticipation cost is estimated to be £8.8 billion, and this section summarises the costs.

Defensive expenditure – buildings

Defensive expenditure in buildings includes any cost incurred from measures to improve fire safety and protect against fires in buildings (such as fire doors and compartmentation structures). It includes the cost of incorporating active and passive fire protection in buildings, and from designing buildings in a fire safe way. To calculate this cost, a top-down method based on ONS construction expenditure and market size data was used. Three different elements are included, which give a total cost estimate of £4.6 billion.

Fire safety costs in new constructions

Calculated as 1.25% of the total construction costs for residential and 5% all other construction. Construction costs are based on ONS estimates of England construction costs. Calculated by multiplying ONS estimates of England construction costs by the proportion of these costs estimated to be fire related (1.25 % of residential buildings and 5% in all other construction (Ashe et al., 2009, further details in chapter 3). Total cost estimate: £3.5 billion.

Fire safety costs in construction repairs and maintenance

Calculated by making similar adjustments to ONS data and assumptions on fire related expenditure (Schaenman et al., 1995, further details in chapter 3). Total cost estimate: £793 million.

Ongoing maintenance of fire safety equipment and measures

Calculated as approximately 8% of the total cost of measures in new construction (Schaenman et al, 1995 and Ashe et al., 2009). Total cost estimate: £284 million.

Defensive expenditure – consumer goods

There is a cost to companies incurred when the consumer goods they sell (such as electrical appliances, furniture, and motor vehicles) are required to meet fire safety standards and comply with the appropriate legislation to ensure they are fire safe. This may involve the cost of fire safety testing, adding protective markings, applying fire retardants, or using more expensive less flammable or combustible materials. This represents an additional cost to businesses, which in many cases is built into the cost of consumer items and passed on to consumers.

As there is no available estimate for the size of the consumer goods fire prevention market, the cost was calculated using ONS estimates for UK manufacturing sales in numerous industries assumed to require some level of fire safety in their goods (such as electronic products and electrical goods or equipment) and the available literature. Numerous studies were assessed, which found that the proportion of manufacturing output that is defensive expenditure on consumer goods is approximately 1.5% to 2.5%. The lower bound assumption of 1.5% was taken in this report, giving an overall cost of £2.0 billion.

Carbon emissions from fixed fire suppression systems

Some commercial buildings have fixed fire suppression systems that aim to prevent fires that may have required FRS attendance. These lead to costs in terms of the release of Hydrofluorocarbon (HFC) gas. This gas is generally used because it is a method of extinguishing fires without damaging electronic equipment, so is often present in computer rooms, data centres and telecommunication facilities. The National Atmospheric Emissions Inventory (NAEI) release annual estimates for the total carbon emissions from these systems. This was multiplied by the Department for Business, Energy and Industrial Strategy BEIS unit cost of carbon (£228 per tCO₂e in year ending March 2020 prices) to give a total cost of £60 million.

Insurance administration

The expenses incurred from administration of developing insurance policies, and the running costs of insurers to facilitate the delivery of these policies, is considered to be an economic cost to society because these resources could have otherwise been used in alternative productive activities. The costs are calculated using data from the Association of British Insurers (ABI) to give a total cost of £387 million.

Fire training and awareness in workplaces

The possibility of fires occurring in workplaces means that many individuals need to undertake workplace fire safety activity, which has an economic cost in terms of time spent. This cost is calculated using ONS data on the number of individuals in employment and their employment cost, multiplied by an estimate of the amount of time they will spend on fire training per year (estimated to be 45 mins per individual per year, acknowledging that training levels will vary). Total cost: £336 million.

Expenditure by fire and rescue services

The total FRS budget was approximately £2.2 billion in year ending March 2020, which includes funding their response to fire, non-fire activity, and activity in anticipation of having to deal with fires, which includes training, resilience requirements, and any activity that aims to prevent and protect against fires occurring. Analysis was completed to split FRS spend into response and anticipation activity (which included all time and cost not on direct response), and then between fire and non-fire related spend. This is discussed fully in the response chapter. Total cost: £1.4 billion.

Wider anticipation costs

There are wider anticipation costs referenced in this report, including the cost of defensive expenditure from arson spend, referrals from FRS home fire safety checks (HFSCs) and additional central government activity on fire safety. Some of these costs are monetised, and they are all further discussed in the anticipation chapter.

Total anticipation costs

Table M3 shows the total anticipation costs of fire. The largest anticipation cost comes from defensive expenditure in buildings, with other substantial costs being defensive expenditure in consumer goods and FRS spend. Together, total anticipation costs equal £8.8 billion.

Table M3: Total cost in anticipation of fire

Fire type	Total cost
Defensive expenditure - buildings	£4,626m
Defensive expenditure - consumer goods	£1,990m
Defensive expenditure - arson	£11m
Carbon emissions from suppression systems	£60m
Time cost of training in businesses	£336m
Insurance administration	£387m
FRS expenditure	£1,384m
FRS partner HFSC	£0.3m
Total costs	£8.8bn
Unit cost (all fires - 154,109)	£57,100
Unit cost (primary fires - 68,765)	£127,900

Table M3 notes:

1. All total costs rounded to an appropriate value for full transparency, and unit costs to the nearest £100.

2.2 Consequence

Consequence costs are defined as the direct and indirect costs that occur as a result of fires that occurred in year ending March 2020, such as property damage, loss of business, human injury, and fatalities. Table M4 gives a breakdown of the total consequence cost. Total consequence costs are estimated to be £3.2 billion, and this section summarises the total costs.

Physical and Emotional harms

Fire-related physical and emotional injuries have a cost. This has been calculated using the Quality Adjusted Life Year (QALY) method. The current monetary willingness to pay (WTP) value for a QALY is £70,000 in year ending March 2021 prices (HMT, 2022). From research conducted on the QALY impact of injuries (Salomon et al., 2015, GBD, 2019), the duration of these injuries, and data provided from the IRS on injuries resulting from fire incidents, it has been possible to monetise the cost of fire related injuries. Where the length of an injury exceeded a year, total costs were discounted at a 1.5% annual rate (HMT, 2022). The IRS records different severities of injury that were used to weight the duration and QALY impact. Where literature regarding the impact of a specific injury was not available, appropriate proxies were taken. These proxies were the best assumptions available in these cases where data on the impact of specific injuries was not available. Emotional injuries were calculated as above, though with an extra variable for the likelihood of experiencing emotional trauma. The emotional harm from rescues was calculated using a similar method, but with a reduced unit cost of harm and by using the number of rescues.

The QALY impact of fatalities was also calculated using IRS data on the number of fire-related fatalities, the sex of the individual and their age at the time of death. To calculate the years of life lost from a fatality, the age at death was subtracted from expected years left of life (ONS, 2021c). The years of life lost were then multiplied by the QALY and total costs were discounted at a 1.5% annual rate (HMT 2022) to put the costs into present value terms, and account for time preference and systemic risk. Cost of all physical and emotional harms: £362 million.

Property damage

Property damage resulting from fires was broken down into 2 areas: (i) direct property damage and (ii) wider property damage. Direct property damage was calculated by taking the estimated damage area of fires (in m²) from the IRS and multiplying it by the cost to rebuild that area (per m², Nimblefins, 2022). The IRS records fire damage and total damage, with the latter consisting of both fire damage and smoke and water damage. When calculating direct property damage, fire damage is multiplied by the unit rebuild costs, whereas smoke and water damage is calculated based on unit rebuild costs multiplied by 20%, as smoke and water damage is unlikely to require a full rebuild of space. This 20% downrate factor for smoke and water damage is based on previous assumptions used by Centre for Economics and Business Research (CEBR), 2014.

Wider property damage cost includes impacts such as damage to contents and has been calculated by applying an uprate factor to the direct damage figure. For dwelling fires, wider property damage is assumed to be 28% of direct damage, whereas for other building fires, it is considered to be 133% to account for the greater potential for business impacts (Fire Protection Association, 2020). Cost: £2.0 billion.

Road Vehicle damage

Damage from road vehicle fires was calculated using the 14 vehicle categories within the IRS (excluding bicycles), and by equating damage categories (such as fuel tank and engine compartment) to an estimate for the percentage of a vehicle damaged. An average price for each vehicle type was found, this was then multiplied by the corresponding damage percentage to give a cost per incident. This was completed for all year ending March 2020 incidents, giving a total cost: £272 million.

Lost output

Lost output was split into 4 areas:

1. Property damage / business closures.
2. Time off work following injury.
3. Reduced productivity following injury.
4. Fatalities.

Lost output resulting from business closures and property damage was calculated by estimating the number of employees rendered unemployed or unable to work as a result of a fire, and the lost earnings to an employee are assumed to represent the lost output to a business. The CEBR, 2014 paper, which estimates the area of damage required to render an individual unemployed in warehousing, was used as a basis for these calculations. By comparing the employee density (Homes and Communities Agency, 2015) of warehousing to 7 other industries (by building categories within the IRS) it was possible to find the damage area required per fire to lead to an individual being unemployed by industry. This was then adjusted to account for how likely it is that individuals could work from home, or not in the workplace location. Total damage area was used, as smoke and water damage would also impact the ability of a business to operate. The number of unemployed by industry was then multiplied by the average yearly salary (by industry), and the estimated duration of unemployment (ONS, 2022b), which was adjusted to account for different labour market flexibilities. This gave total costs by industry, which were then summed. This method assumes that salary is equal to marginal product of labour. It was not possible to monetise returns to capital or producer surplus due to a lack of available data.

To calculate the cost of time off work following an injury, the 4 severity categories within the IRS injury day were used to estimate duration. A literature review was conducted on the average length of time it takes an individual to return-to-work following an injury, by injury severity, with an assumption of 5.5 hours for precautionary checks and first aid required, 5 weeks for hospital admissions (slight) and 17 weeks for hospital admissions (severe). These durations were multiplied by number of injuries and the average monthly employment rate in the year ending March 2020 (ONS, 2022a).

Reduced productivity following injuries used the same QALY impact method as physical harms, though used average yearly salaries instead of the QALY value. The QALY impact of the injury was used as a proxy for decrease in productivity, and this was multiplied by duration, employment rate and total hours worked in a year (1,924) (ONS, 2022d) to find overall lost productivity.

Lost output from fatalities were calculated in the same way as in harms, though instead of life expectancy, years left until retirement were used, with those over 65 excluded. Cost: £170 million.

Health services

Health service costs are based on NHS unit costs for different treatments, with assumptions made on what the likely treatment is for each injury based on IRS categorisation of injury type and severity. Only incidents recorded as a hospital admission are included. Physical harms also factor in ambulance and medical procedure costs (NHS, 2021) and emotional harms are calculated using counselling costs (Curtis and Burns, 2020). Cost: £23 million.

Environmental

Three different cost areas were included in this section:

1. Carbon cost of emissions.
2. Impact on air quality.
3. Damage from outdoor fires.

The carbon costs were calculated using BEIS data on total carbon emissions and the unit cost of carbon emissions. The air quality costs were calculated using estimates for the total volume of particulate matter emissions and the Department for Environment, Food and Rural Affairs (DEFRA) unit costs, and the damage from outdoor fires was calculated using 4 case studies (referenced and detailed in section 4.5.3) on the cost of cleaning up and restoring damaged land. Cost: £123 million.

Insurance administration

The expenses incurred administration and management of fire damage claims is considered to be an economic consequence cost to society because these resources could have otherwise been used in alternative productive activities. The costs are calculated using data from the Association of British Insurers (ABI) to give a total cost of £114 million.

Wider costs

The cost of crime (arson) related impacts (including Police, CJS and victim services) are estimated using uprated unit costs from the economic and social costs of crime report (HO, 2018). Cost: £78 million.

Firefighter injuries are costed in the same way as injuries to other victims of fire, using IRS data on injury type. Cost: £0.4 million.

Non-monetised costs

There are several consequences of fire which could not be monetised to a robust enough degree for inclusion within the headline overall consequence estimates. These non-monetised impacts are described in section 6 and include the cost of evacuations because of fire and the fear of fire. Further work could look to monetise these costs.

Table M4: Summary of consequence costs

Cost area	Total cost	Unit cost (all fire incidents)	Unit cost (primary fires)
Physical and emotional harms	£362m	£2,400	£5,300
Physical harms	£39m	£300	£600
Emotional harms	£7m	£40	£100
Fatalities	£314m	£2,000	£4,600
Rescues	£2m	£10	£30
Property damage	£2,009m	£13,000	£29,200
Vehicle fires	£272m	£1,800	£3,900
Lost output	£170m	£1,100	£2,500
Lost productivity (harms)	£10m	£60	£140
Lost output (harms)	£7m	£40	£100
Lost output (fatalities)	£35m	£200	£500
Lost output (damage)	£119m	£800	£1,700
Health service costs	£23m	£150	£300
Health service costs (physical)	£21m	£140	£300
Health service costs (emotional)	£1m	£10	£10
Ambulance & fatalities	£1m	£10	£20
Environmental harm	£123m	£800	£1,000
Insurance	£114m	£700	£1,700
Wider costs	£78m	£500	£1,100
Arson associated costs	£78m	£500	£1,100
Firefighter injuries	£0m	£0	£10
Total	£3.2bn	£20,500	£45,000

Table M4 notes:

1. All total costs rounded to nearest £0.1 billion or £1 million and all unit costs rounded to the nearest £100, or £10 if under £200.
2. Based on 154,109 incidents, and 68,765 primary incidents. Unit costs calculated by dividing total costs by these numbers, except for the environmental primary fires figure which uses the total cost associated with primary fires only (£68m).
3. Secondary and chimney fires have no expected consequence costs (except for some environmental costs), so unit costs with and without these incidents are included.

2.3 Response

FRSs spend time responding and dealing with fires when they occur. This incurs an economic cost, which includes the opportunity cost of FRS time spent responding to incidents (this time could have been used on other productive activities), the direct cost of non-labour expenditure related to incident response, and the cost of control staff (who facilitate and co-ordinate responding fire appliances). The cost of response only considers direct costs incurred as a result of fire incidents, and so is not equivalent to total FRS expenditure. The wider costs of resilience – training, so staff can effectively respond to incidents, and other FRS activities (such as prevention and protection activity) are not classed as response related, and so are not included in the FRS cost of response. These costs are instead included in the FRS anticipation section of this report.

Table M5 shows the total cost of responding to fire. The cost of FRS response to fires is calculated to be £73.9 million in year ending March 2020, which includes the cost of direct labour (firefighters), direct non-labour expenditure (including fuel and appliance wear and tear), and control costs. The costs were calculated using a crewing survey, which was sent to all FRSs in 2021 and requested data on average appliance crewing (by staff role), average non-labour expenditure as a direct result of incidents (such as fuel and appliance wear and tear), and the average salary of control and firefighting staff. This data was combined with IRS data on time spent by FRSs at incidents, and Home Office (HO) published statistics on the number of control staff, to estimate the direct impact on each FRS from responding to fires. The cost of response varies by FRS, and the total figures are presented in the table below.

Table M5: Total response costs

Total cost of fire response	Total cost	Unit cost (152,289 incidents)
Direct Labour	£30.4m	£200
Direct Capital	£21.7m	£140
Control	£21.8m	£140
Total	£73.9m	£490

Table M5 notes:

1. Some incidents (with no time reported due to IRS issues) not included. All total costs rounded to the nearest £0.1 million, and all unit costs rounded to the nearest £10.

The response section of the report also considers the cost of responding to different incident types such as fire false alarms (£34.9 million) and non-fire incidents (£48.8 million). There is also a detailed outline of how this methodology and these figures are used to estimate the cost of FRS anticipation spend (£1.4 billion), which is included in the anticipation totals in this report. Currently, only the cost of FRS response is included in this section. Further work could be completed to include the response cost of industrial fire brigades, and wider costs associated with response (such as central call centre handling costs). The cost of other agencies responding to fires (such as health service, police and CJS costs), are included in the consequence section of this report.

2.4 Total cost

The total costs of fire, split by cost area, fire type and by unit and total cost, are presented in tables M6 to M8 below. These tables allow for an easy comparison of how costly different fires are, and of what is driving the cost differences. Table M6 shows the total economic and social costs of fire in year ending March 2020, by cost area.

Table M6: Total economic and social costs of fire

Cost area	Total cost	Unit cost (all fire incidents)
Anticipation	£8.8bn	£57,100
Defensive expenditure - buildings	£4,626m	£30,000
Defensive expenditure - consumer goods	£1,990m	£12,900
Defensive expenditure - arson	£11m	£70
Carbon emissions from suppression systems	£60m	£400
Time cost of training in businesses	£336m	£2,200
Insurance administration	£387m	£2,500
FRS expenditure	£1,384m	£9,000
FRS partner HFSC	£0.3m	£0
Consequence	£3.2bn	£20,500
Physical & emotional harms	£362m	£2,400
Property damage	£2,009m	£13,000
Vehicle fires	£272m	£1,800
Lost output	£170m	£1,100
Health service costs	£23m	£200
Environmental harm	£123m	£800
Insurance administration	£114m	£700
Wider costs	£78m	£500
Response cost	£74m	£490
Total	£12.0bn	£78,000
Direct/Marginal cost	£3.2bn	£20,900

Table M6 notes:

1. All total costs rounded to an appropriate value for full transparency, and all unit costs rounded to the nearest £100, unless shown otherwise.

Table M7 focuses on the total direct marginal cost of fire and shows it by cost area and fire type. There are large differences in cost between some areas of fire.

Table M7: Total direct marginal cost of fire by fire type

Cost area	Total cost	Incidents	Unit cost (all fire incidents)
Consequence	£3,152m	154,109	£20,500
Of which is primary fires	£3,097m	68,765	£45,000
Other buildings	£1,755m	14,328	£122,500
Dwellings	£901m	28,491	£31,600
Other Outdoors	£82m	5,386	£15,300
Road Vehicles	£358m	20,560	£17,400
Secondary (total)	£55m	85,344	£600
Response	£74m	152,289	£500
Of which is primary fires	£59m	67,831	£900
Other buildings	£23m	14,129	£1,600
Dwellings	£21m	28,092	£800
Other Outdoors	£8m	5,295	£1,400
Road Vehicles	£7m	20,315	£300
Secondary (total)	£15m	84,458	£180
Direct/Marginal cost	£3,226m	-	£20,900
Of which is primary fires	£3,155m	-	£45,900
Other buildings	£1,779m	-	£124,200
Dwellings	£922m	-	£32,400
Other Outdoors	£90m	-	£16,700
Road Vehicles	£364m	-	£17,700
Secondary (total)	£70m	-	£800

Table M7 notes:

1. All total costs rounded to an appropriate value for full transparency, and all unit costs rounded to the nearest £100, except for those under £200.

Number of in-scope incidents differ between consequence and response due to the removal of a small number of incidents attended with errors in the mobilisation data (for example, with mobilisation times missing or deemed incorrect).

Table M8 shows the total direct marginal cost of fire, split by accidental and deliberate fires. Annex table A2 shows total primary marginal cost for accidental and deliberate fires, split by fire type.

Table M8: Total direct marginal cost of fire by accidental and deliberate fires

Cost area	Total cost	Unit cost (all fire incidents)	Unit cost (primary incidents)
Consequence	£3.2bn	£20,500	£45,000
Accidental	£2.2bn	£26,300	£44,000
Deliberate	£0.9bn	£13,400	£47,700
Response	£74m	£500	£900
Accidental	£53m	£600	£900
Deliberate	£21m	£300	£700
Direct/Marginal cost	£3.2bn	£20,900	£45,900
Accidental	£2.3bn	£26,900	£44,900
Deliberate	£1.0bn	£13,700	£48,400

Table M8 notes:

1. All total costs rounded to nearest £0.1 billion or £1 million. All unit costs rounded to the nearest £100. Total may not sum due to rounding.

3. Costs in anticipation of fire

Anticipation costs are the most challenging cost area in this report to fully define. Often they are difficult to monetise with available data, and they will not, in the short term, be directly impacted by changes in the number of, or consequences from, fires. For the purposes of this report, anticipation costs are defined as expenditure incurred in year ending March 2020 on measures to prevent or protect against fires.

Historical expenditure on items still in place in year ending March 2020 that protect against or prevent fires happening (such as the purchase of a fire extinguisher in year ending March 2019, that is still in a building, or fire training for an individual that is still remembered) is not included in this report. These costs are unchangeable and no action or change in fire numbers could reduce or prevent them, so it is not accurate to include them in the total cost of fire for the year year ending March 2020. It is also acknowledged that there will be benefits of fire anticipation cost from year ending March 2020 that will occur in future years, however these impacts are also not considered in this report. Some costs (such as fire training and carbon emissions) included are not financial expenditure but wider economic costs that were incurred in year ending March 2020.

Anticipation costs are included in the unit costs of fire, however, they are not included in the marginal costs of fire, as in many cases anticipation costs may not fall proportionately to the number of fires. A level of protection and prevention expenditure, unlike consequence costs, will likely always be desirable and incurred to prevent against the risk of fires, even if the number of fire incidents was to fall to extremely low numbers^{[footnote 2]}. Some of the costs monetised in this section arise from government regulations, which impose additional costs and burdens on businesses and individuals. It is beyond the scope of this report to assess the societal benefits of these regulations and of incurring this cost.

This section of the report covers the cost areas of:

• defensive expenditure in buildings
• defensive expenditure in consumer goods
• defensive expenditure against arson
• carbon emissions from fixed fire suppression systems
• insurance administration
• fire training and awareness in workplaces
• expenditure by fire and rescue services (FRSs)
• wider anticipation costs
• total anticipation costs

3.1 Defensive expenditure - Buildings

There is a cost of incorporating active and passive fire protection in buildings, and from designing buildings in a fire safe way. In basic terms, active fire protection refers to measures that aim to detect and directly stop fires (such as fire alarm or sprinkler systems) and passive fire protection refers to measures aimed to contain and reduce the spread of fires (such as fire doors and effective compartmentation). The cost of these measures in buildings are widely included in cost of fire reports in other nations^{[footnote 3]}, and often represents the largest single cost due to the scale and range of buildings covered. This cost aims to capture the cost of abiding by relevant fire safety legislation, including the cost on responsible persons in buildings covered by the Regulatory Reform (Fire Safety) Order 2005 (the FSO)^{[footnote 4]}, and building regulations for new construction.

There is no single way of calculating this cost currently, and it is a difficult area to accurately assess. We are not aware of any recent, accurate, published UK market research reports that provide a comprehensive assessment of this cost. To monetise the cost, we have used a top-down method based on Office for National Statistics (ONS) construction expenditure. This considers 3 areas, new construction, building repairs and maintenance and the ongoing maintenance of fire safety equipment and measures. It is not possible to split the cost into those from legislative requirements versus fire safety standards, or to count “actual costs” incurred, so a top-down assumption is made. The additional cost of unsafe goods, and the risk they pose, has not been monetised.

3.1.1 Fire Safety costs in new constructions

The ONS publishes total annual estimates for construction output in Great Britain, by building sector (ONS, 2021a). These were adjusted to solely consider construction in England using the proportions of total output by construction type by region (ONS, 2021a). This gave a total construction cost of £155.0 billion in England in year ending March 2020, of which £101.4 billion was related to new construction work.

The assumption for fire safety costs, as a proportion of total expenditure on construction, is based on numbers published within Ashe et al., (2009) for the cost of fire protection as a percentage of construction costs in the UK: it is assumed that 1.25% of spend in residential buildings, and 5% of the spend in all other construction (including infrastructure, industrial and commercial spend) is incurred as fire defensive expenditure. This covers active and passive fire protection, includes the cost of abiding by fire safety legislation on materials used and building design, and any additional work that occurs in the building to make it fire safe. Ashe et al., (2009) is the only known analysis of this type specific to the UK and therefore the best available source for this analysis^{[footnote 5]}. Further detailed work could be undertaken to improve these assumptions, and this is further discussed in chapter 6: Future developments. It is possible that changes in regulation since this study will have led to a higher proportion of spend being related to fire regulation. Using this methodology and assumptions, the total cost estimate for fire safety costs in new build constructions is £3.5 billion. This cost can be split out using the ONS construction figures into defensive expenditure related to dwelling fires (£506 million), and defensive expenditure related to other building fires (£3.0 billion).

3.1.2 Fire safety costs in building repairs and maintenance

Similar to the cost of defensive expenditure embedded in new building constructions, there is also an element of fire safety costs embedded in repairing and maintaining buildings over time, and the updating of fire protection in buildings when these repairs and maintenance occur.

The ONS publishes figures on annual construction spend on repair and maintenance in housing and other construction. The total value of this is less than total new build construction, with a current estimate of £53 billion in England in year ending March 2020^{[footnote 6]}. There is limited evidence available on what proportion of this spend is related to fire safety and installing active or passive fire protection. So, assumptions from Schaenman et al., (1995)^{[footnote 7]}, the seminal work in this area and only known analysis of this type, is used. It is assumed that 1% of spend in housing (both public and private) is spent on fire safety, and 2% of spend in non-housing repair and maintenance construction. These figures are lower than those on new building construction because repairs are more often made to features without a fire protection element as opposed to those with a fire protection element when compared to overall building construction. In addition, many repairs may be minor changes or improvements, and so only some element of repairs have a fire safety component^{[footnote 8]}.

Further detailed work could be undertaken to improve this assumption. Using this methodology and assumption, the total cost of fire protection in construction repairs and maintenance is estimated to be £793 million. This cost can be split out using the ONS construction figures into defensive expenditure related to dwelling fires (£271 million), and defensive expenditure related to other buildings fires (£522 million).

3.1.3 Ongoing maintenance of fire safety equipment and measures

As well as the cost of new fire protection and fire safety measures related to construction, repairs and maintenance, there is an ongoing cost of directly maintaining fire protection measures. This is separate to expenditure related to building repairs and maintenance, as this only refers to the maintenance of fire safety measures, not the building itself. To give examples, the maintenance or repair of chimneys, fireplaces or walls in a building would fall within 3.1.2, as these are construction related repairs which have an associated fire safety or protection costs to them. Whereas the cost of replacing or maintaining specific active or passive fire protection systems such as smoke alarms, fire doors or sprinkler systems separately to building repairs do not have an associated construction cost, so are not captured in 3.1.2, and are instead captured in this cost. The costs associated with testing and inspecting fire protection systems are also assumed to be part of this estimate. Due to the top-down approach of this methodology, it is not possible to be certain that there is no double counting in these figures, or split this cost out into specific maintenance activities. However, the clear definitional difference is used to mitigate this issue. These costs fit within the definition at the start of this section, as they are “expenditure in year ending March 2020 on measures to prevent or protect against fires which could happen in the same year”. Therefore, they fit within the flow of anticipation costs, not the stock. They cannot be considered sunk costs because individuals could likely choose not to repair or maintain their fire protection measures without incurring any costs^{[footnote 9]}.

It is assumed that the maintenance costs of fire safety equipment and all active and passive fire protection measures is approximately 8% of the total cost of measures in new construction, which is the best available estimate for this proportion (used in both Schaenman et al., 1995 and Ashe et al., 2009). This estimate only considers the maintenance of relatively new systems, with both papers arguing that this is likely to be the estimate for what new fire safety measures would likely cost on an ongoing basis as soon as they are installed, or from the year after installation. This means that the cost is unlikely to fully consider the maintenance and repair costs of all previous generations (defined as annual spend on fire protection systems) within buildings. However, as you go further back in time, it becomes harder to estimate the annual cost of maintaining equipment; especially as many of these will have likely been replaced with new building construction, or alongside wider construction repairs and maintenance, meaning their costs would fall within section 3.1.1 or 3.1.2 of this report. To prevent potential double counting and overestimating the size of this cost, a conservative assumption of 8% has been applied to new build construction costs, which aims to fully capture the cost of defensive expenditure in buildings and construction. This cost is estimated to be £284 million. This cost can be split out into defensive expenditure related to dwelling fires (£41 million), and defensive expenditure related to other buildings fires (£243 million).

3.1.4 Total costs

The total cost of defensive expenditure in buildings and construction is calculated by summing these 3 elements of cost to reach an estimate of £4.6 billion. This is considered to be the total cost of the active and passive and protection industry, and include the cost of all defensive expenditure, alongside all impacts on responsible persons as defined by the FSO. This cost can be split out into defensive expenditure related to dwelling fires (£0.8 billion), and defensive expenditure related to other building fires (£3.8 billion).

3.1.5 Evaluation of method

As the assumptions within this cost estimate are relatively uncertain, and it is recognised that further detailed work would be required to fully estimate these costs, a number of alternative methods have also been completed that aim to provide a sense check for the total costs identified in section 3.1.4. These methods are not considered preferable to those in sections 3.1.1 to 3.1.3, hence are only included in this section of the report. They have notable caveats and are not directly comparable to the total estimate in 3.1.4 as they often do not include the full cost of defensive expenditure, and only focus on protection measures in buildings. They have been included for full transparency, and to provide greater assurances on the accuracy of the £4.6 billion total figure.

Evaluation 1: Uprating figures from the Department for Communities and Local Government (DCLG) 2011 report (DCLG, 2011a, DCLG, 2011b)^{[footnote 10]} and previous studies on the UK fire protection market.

In 2011, DCLG published estimates for the cost of fire in the UK (DCLG, 2011a, DCLG, 2011b), which were based on a research document by the Fire Industry Association (FIA) entitled “Survey of the UK active and passive fire protection markets”. This report estimated the total value of the market in the UK including product costs, installation costs and maintenance costs. The most recent estimates this report has published are from 2006, in which they estimated that the total size of the UK market was £2.965 billion (BRG, 2009). Converting this to an equivalent figure for just England gives a cost estimate of £2.57 billion^{[footnote 11]} in 2006.

To uprate this estimate to a 2019 equivalent figure, 3 scenarios were modelled for how the UK fire protection market may have changed since 2006.

(i) The first scenario was to assume the growth rate for the market mirrored changes in construction output.

Using ONS construction data, it could be calculated that the construction output had grown at approximately 3.1% annually between 2006 and 2019 (ONS, 2021a), meaning the fire protection market in England would be valued at approximately £3.8 billion in year ending March 2020.

(ii) The second scenario modelled was to assume that the fire protection market continued to grow at the same rate as it had before the report was published.

The market had grown at an average rate of 4% between 2001 and 2006 (BRG, 2009). Extending this trend forward gives an estimate of £4.3 billion for the size of the 2019 market.

(iii) Finally, the third scenario modelled is that the growth rate may have exceeded previous estimates years and been approximately 5% annually over the 13-year period (Roofing Today, 2018, AMA Research, 2021, Plimsoll, 2021)^{[footnote 12]}.

In this scenario, the total 2019 market size would be approximately £4.9 billion.

The growth rate of the fire protection market over the last 13 years is uncertain, which is why this method only acts as a way of evaluating the total costs in 3.1.4 and is not the central method used in this report. As our estimate of £4.6 billion falls within this estimated range of £3.8 to £4.9 billion, this evaluation supports the figures used in this report. An annual increase in the market of 4.6% would make the DCLG figures consistent with the ones we have calculated, which is considered a fair, and potentially conservative growth rate assumption. It is also expected that our estimate would fall on the higher end of this cost range, because our definition for defensive expenditure is broader than the one used in the FIA survey (ours considers material and responsible person costs).

Evaluation 2: Comparison of figures to estimates for the global protection market.

An alternative method to estimate the cost of defensive expenditure in buildings and construction in England is to compare our estimates to those put together in international studies on the size of the global protection market. These comparisons are difficult due to differences in methods used and difficulties in estimating the proportion of the global market, that is England specific, yet it remains a useful sense check. The global fire protection market is estimated to be approximately $92 to $97 billion (Vision Research Reports, 2020, Allied Market Research, 2020) in 2019, of which approximately 35% is North America ($32 to $34 billion) (Vision Research Reports, 2020, Allied Market Research, 2020). This figure can be converted to a value for England by altering the estimate to US specific figures (using the GDP of the North American nations), exchanging the estimate to pounds sterling using purchasing power parities (PPP) (OECD, 2023), and then adjusting for differing populations in the US and England^{[footnote 13]}. This gives an estimate that the English fire protection market is approximately £3.3 to £3.5 billion. This is significantly lower than our proposed estimate, highlighting some of the uncertainty in our analysis, whilst still acknowledging significant limits in converting international estimates to England equivalent ones. The built environment, and fire safety regulations, will differ across countries, so it is not expected that these costs would directly align with our proposed estimates. It is also expected that the global protection market figures do not include all costs incurred in construction, or include the full cost of infrastructure construction, and so it is expected that they underestimate total costs compared to our definition.

The method we use in this report is considered to be the best available with the current research available. The methodology does not fully consider changes to the built environment, and regulatory changes in recent years that could significantly increase the cost of defensive expenditure in buildings. The removal of ACM cladding, and remediation in medium and high-rise buildings to make them more fire safe, will likely represent a significant additional cost that is not fully included in this report, and will not be accounted for using general market trend analysis. These changes may also mean that the assumptions used in 3.1.1 to 3.1.3 underestimate the full defensive expenditure cost. Future work could look to fully estimate this cost, and would include government, business, and individual expenditure. Estimating defensive expenditure by building type, was not possible due a lack of available data.

3.2 Defensive expenditure - consumer goods

Companies incur costs when making the consumer goods (such as electrical appliances, furniture, and motor vehicles) they sell meet fire safety standards and comply with the appropriate legislation to ensure they are fire safe. This may involve the cost of fire safety testing, adding protective markings, applying fire retardants, or using more expensive less flammable or combustible materials. This represents an additional cost to businesses, which in many cases is built into the cost of consumer items and passed on to consumers. As this cost is related to fire safety and prevention, it is a societal cost that should be included within the anticipation section of this report. Appliances that are not fire safe and do not comply with appropriate fire safety legislation or standards are generally likely to be cheaper to produce, albeit illegal to sell. This suggests that there is a cost incurred to ensure that goods are fire safe and comply with the appropriate fire safety legislation or legislation. Due to the size of the consumer goods industry, this cost is potentially large and so must be considered within this analysis. Many fires each year are caused by faulty electrical appliances. In year ending March 2020, there were 3,874 domestic fires caused by faulty appliances and leads (HO, 2023, FIRE0601) and consumer goods. The existence of fire safety standards and legislative requirements in consumer goods, which reduces this number, has large potential benefits in terms of a reduced prevalence of fires, fire related injuries and fatalities. However, the calculation of these benefits is beyond the scope of this report and so has not been monetised.

There is no publicly available known estimate for the size of the consumer good fire prevention market that could be used within this analysis. Therefore, novel analysis has been completed to estimate the cost of ensuring consumer goods are fire safe, based on the size of the consumer goods market and the proportion of costs that are related to fire safety.

The ONS publish annual estimates for total UK manufacturing sales by product (ONS, 2020a), calculated using a ProdCom survey. In 2019, total manufacturing sales in the UK were estimated to be approximately £402 billion. These sales were examined by industry type and only the manufacture of goods with an expected fire safety requirement were included in the analysis. This included the cost of manufacturing electronic products, electrical goods, machinery, electrical equipment, furniture, and motor vehicles. The total manufacturing sales when only including these industry types is estimated to be £167 billion.

The next step is to estimate what proportion of the cost of these manufacturing sales is related to fire safety. This is novel analysis, with high levels of uncertainty due to the potential variation by industry and type of good. Some goods will require higher expenditure to make them fire safe than others and there is limited literature on the proportion of the cost of goods that is related to fire safety. The seminal work in this area, Meade (1991), calculated that 30% was a reasonable overall factor for fire-grade design, which was applied to consumer goods with an electrical aspect. These estimates have been built on in many more recent cost of fire reports, in Canada, Australia and the US (Schaenman et al., 1995, Ashe et al., 2009, Zhuang et al., 2017) although in some of the reports these have been identified as overestimations, potentially being a factor of 2 to 10 times too high (Richardson, 1995). The total cost has been estimated in a number of studies, ranging from USD$54 billion in the US National Fire Protection Association (Zhuang et al, 2017) report, to AUD$1.6 billion in Ashe et al., (2009) on Australia, to CAD$2 billion in Canada (Schaenman et al., 1995). These studies can be used to estimate what proportion of the cost of manufacturing sales is related to fire safety. Over the studies, the proportion of defensive expenditure to total manufacturing output is relatively constant, at approximately 1.4% to 2.5%^{[footnote 14]}.

This offers us a starting point for the assumption. To prevent overestimating the size of this market, we only focus on the manufacture of goods with an expected fire safety requirement (as opposed to all manufacturing output) and take the lowest bound of these estimates, that 1.4% of the cost of these consumer goods is related to making them fire safe^{[footnote 15]}. This is a highly uncertain assumption, which will vary by industry and consumer good types, however it is seen as the best initial assumption for the overall cost in consumer goods and is broadly consistent with other studies. Due to this uncertainty, estimates for each industry are not calculated separately.

Logically, it can be inferred from this assumption that there would be a 1.4% reduction in the cost of production if all fire safety standards, legislative requirements, and testing were to be removed. Assuming these cost changes are passed through to consumers, there would also be a 1.4% reduction in the price of these goods.

Applying the 1.4% conversion factor to the £167 billion cost of in-scope consumer goods in the UK, and then downrating by the proportion of the UK population that live in England (84%) (ONS, 2020b) to account for only the English market being in-scope of this report, gives a total cost of fire safety in consumer goods of approximately £2.0 billion in year ending March 2020.

For full transparency, different scenarios are presented. If the high estimate in the literature of 2.5% was taken, the cost of fire safety in consumer goods would rise to £3.5 billion. Alternatively, if an assumption of one per cent was taken, the cost would fall to £1.4 billion. It was not possible to split this cost out by fire type due to a lack of data on the cost and location of consumer goods. Further work could be completed to include this in future analysis.

When compared to studies in other nations, this is a conservative estimate for total cost. There is currently very limited published evidence on how the cost of fire safety regulations differs between the nations modelled in this report (Canada, Australia and the US) and the UK. Therefore, this analysis presents an initial starting point to estimate this cost, but further work, for example market research, is needed to accurately cost this expenditure.

The full cost of defensive expenditure in consumer goods would be the total burden incurred by English manufacturers and consumers. In this report, the cost is proxied using English production data, and there is no consideration of how much of the burden falls on consumers instead of producers. If we were to assume that at least some cost was passed through to consumers, it is likely that some of these costs should not be included in this report because they are incurred by consumers outside of England (when goods are exported). However, equally, it may mean that the cost is underestimated because costs incurred from imports, that are passed on from non-English manufacturers to English consumers, are not included. Difficulties in getting international and intra-UK trade data that directly aligns with the ONS data, uncertainties in the costs incurred by foreign manufacturers, and uncertainties on the elasticity of each type of consumer good (which will determine how much of the cost is passed onto consumers) mean that this is not considered in this analysis.

Product recall is an important impact to consider when assessing the cost to businesses of ensuring their goods are fire safe, as businesses could potentially incur large costs if their products must be recalled. Some datasets exist that show the goods currently subject to recall^{[footnote 16]} however, there is no known dataset that includes the number of goods that are recalled, and the cost of this on individuals and businesses. The lack of this data means that it is incredibly difficult to ascertain the cost impact of product recall on businesses, so it has not been costed separately in this report.

3.3 Defensive expenditure – arson

Defensive expenditure in the context of the economic and social cost of crime is the money spent by individuals on crime detection and prevention. This encompasses estimates for both security equipment (such as burglar alarms, CCTV equipment, car alarms) and private security (such as door supervision). In the context of fires, this covers all spend by individuals to protect themselves from arson offences. The methodology adopted for this uses the same approach as the economic and social cost of crime report (HO, 2018), updated to account for inflation using the HMT GDP deflator (HMT, 2023).

The full methodology for this cost is presented in the published cost of crime report, with the total arson related defensive expenditure in England^{[footnote 17]} being estimated to be £11 million.

3.4 Carbon emissions for fixed fire suppression systems

A further anticipation cost of fire is carbon emissions resulting from fixed fire suppression systems in commercial buildings. These fall within anticipation as they prevent or protect against fires that would have otherwise required additional FRS response action^{[footnote 18]} and therefore led to response, and potentially, consequence costs. The National Atmospheric Emissions Inventory (NAEI) in BEIS^{[footnote 19]} publish annual estimates for the total carbon emissions from “firefighting” (BEIS, Table 1.2, 2021b). In this context, “firefighting” refers to the release of Hydrofluorocarbon (HFC) gas from fixed fire suppression systems in commercial buildings. This gas is generally used because it is a method of extinguishing fires without damaging electronic equipment, so they are often used in computer rooms, data centres and telecommunication facilities that have a large amount of this equipment.

This data estimates that the total volume of emissions from firefighting is approximately 0.31 million tonnes carbon dioxide equivalent (MtCO₂e) per year. As the data covers the whole of the UK, and only England is in scope of this report, the total is multiplied by 84% (ONS, 2020b) to get an updated estimate of 0.26 MtCO₂e. The cost of these emissions can be calculated using the BEIS unit cost of carbon (in year ending March 2020 prices and base year), which is £228 per tCO₂e (BEIS, 2021a)^{[footnote 20]}. This unit cost represents the monetary value that society places on one tonne of carbon dioxide equivalent and is used across government appraisal to allow for an objective, consistent and evidence-based approach to carbon valuation. It is important to note that these unit costs are calculated using a “target-consistent” or “abatement cost” approach and so are related to the cost per tonne of reaching a net zero target, as opposed to the direct harm cost of carbon emissions. The unit costs also increase annually up to 2050, so should be adjusted in line with the BEIS estimates if used for appraisal in different years to 2019. The total cost estimate of carbon emissions in fixed fire suppression systems is estimated to be approximately £60 million in year ending March 2020. All this cost is expected to be related to other building fires as the cost is incurred in commercial buildings.

3.5 Insurance administration

The total cost of insurance claims is not included in this report because it is considered a transfer of money from one party to another as opposed to an overall cost to society. The impact on society that leads to an insurance claim, is instead included in the consequence section of this report (property damage), alongside any administration and management costs from these claims (section 4.6).

However, there are some economic costs incurred from the administration and development of insurance policies, and running costs are incurred by insurers to facilitate the delivery of these policies. These are considered to be an economic cost to society because these resources could have otherwise been used in alternative productive activities (HO, 2018)^{[footnote 21]}.

To estimate the cost of insurance administration in anticipation of fire, data from the Association of British Insurers (ABI) on the total value of insurance company expenses (excluding those on claim management) are used to best estimate the costs directly associated with operating an insurance company, and administrating policies. These expenses include costs associated with preparing and maintaining an insurance contract, and all general administration costs incurred by insurers. This aims to include the full cost of insurance administration (excluding the cost that can be directly attributed to insurance claims, which falls in consequence cost). The ABI data is broken out into domestic and commercial costs, and both are included in this analysis. The data is then adjusted to account for proportion of insurance administration costs which are expected to be fire related, and the amount of market coverage that the ABI have.

Total insurance administration expenditure in year ending March 2020 was estimated to be approximately £387 million.

3.6 Fire training and awareness in workplaces

The presence of fire and the possibility that a fire could occur in workplaces means that many organisations and workplaces, from offices and manufacturing premises to education and health premises need to undertake preventative fire safety work to ensure individuals are aware of what to do in the event of a fire, to ensure their buildings are as safe as possible. Many employers are legally required to provide information, instructions and training to employees related to fire safety in the workplace^{[footnote 22]}. In practice, this will often include specific mandatory training, the completion of fire drills, the reading of fire safety information, and any additional time individuals are required to spend to better understand the threat of fire. This training has a social (opportunity) cost because the time spent undertaking training could have otherwise been spent on alternative productive activities in the workplace.

To estimate the number of individuals who will undertake some level of fire training in the workplace, NOMIS data on the number of individuals aged 16 and over in employment in England in year ending March 2020 is used (NOMIS, 2023). This gave a figure of approximately 27.7 million individuals. Alternative ONS (2022f) data breaks down employment by industry, across the UK, and it was considered whether only industries where specific fire safety training should be included in the scope of these estimates. However, instead it was assumed that all working individuals would need at least some level of fire safety training. Further work could look at how the volume of training varies by job profession and industry to improve these estimates.

Most guidance suggests that workplace fire training should be provided and refreshed at least annually (LFB, 2022), whether this be fire drills, making individuals aware of emergency plans, or more specific training for individuals in the workplace such as fire marshals. To be consistent with the rest of this report, the cost of training should only include the cost of training individuals in year ending March 2020 to prevent fires within that year. As it is expected training would be undertaken annually, this is considered to be a fair assumption. It is likely that training in previous years may still have led to fires being prevented in year ending March 2020 (as there would be an existing stock of training). However, for consistency and simplicity purposes this stock of training is not included in this report.

The level of fire training will vary by industry and job role. Training an individual to become a fire marshal in a workplace with the Institution of Fire Engineers can require a 1 or 2-day course, whereas other employee fire safety training is likely to last from 1 to 4 hours. In other cases, short refresher courses may be required, and in some work premises, individuals may just need to familiarise themselves with fire safety information which may take under 30 minutes to complete. For the purpose of this report, it is assumed that, on average, individuals spend 45 minutes per year on fire safety in the workplace. This figure will likely change by business type and size (with differences in small and large organisations), however splitting the cost out in this manner was not possible with the data available.

Multiplying the total number of in-scope employees with the time spent per individual gives a total time estimate for the time spent on fire safety. This is then multiplied by the average employment cost of £16.16^{[footnote 23]} (ONS, 2022c) to get a total social cost estimate of £336 million, as a result of fire training. As all this cost is related to workplaces, it is all considered to be related to other building fires.

This cost will likely vary by industry depending on number of individuals employed, and amount of fire safety training required. The cost could also be an underestimation because it does not include any wider financial costs of training (such as equipment and employing specific trainers), or the money employers spend on emergency plans and other related fire safety activities. These can be costly in some sectors so could largely increase the cost estimate if included. However, to prevent potentially overestimating this cost, considering the uncertainties already within the current estimate, these costs are excluded from the report.

As this report is based on year ending March 2020 data, it does not consider the impact COVID-19 has had on individuals now working more frequently from home (ONS, 2021b). This could reduce this cost, because individuals will naturally require less or no workplace fire safety training if they are using workplaces less frequently, or not at all.

3.7 Expenditure by fire and rescue services

There are also anticipation costs incurred from activity by FRSs to either directly try to prevent and protect against fires occurring (through activity such as audits and home fire safety checks) or anticipate having to deal with fires (including resilience so being resourced to respond to incidents and being adequately trained). The former cost is considered in some published reports on fire and rescue expenditure and Department for Levelling Up, Housing and Communities (DLUHC) revenue outturn data shows that the cost of community fire safety was approximately £232 million in year ending March 2020 (MHCLG, 2020). This figure is presented for scale but is not the full cost of FRS anticipation spend because it only covers the cost of prevention activity. It does not include the cost of protection staff, training, audits, or enforcement activity, all of which would fall under FRS anticipation expenditure in-scope of this report. There are also some concerns over the accuracy of this data, as not all FRSs complete it in the same way.

Therefore, updated analysis has been carried out which aims to combine DLUHC expenditure and Home Office IRS data to provide an improved view of how to split up total FRS expenditure into fire and non-fire related spend, and then within fire spend, costs related with response versus wider anticipation costs. These would include the cost of community safety activity, training expenditure, organisational readiness, and all back-office functions within services that are important to facilitate FRSs. This method is fully explained in the response chapter of this report, (section 5.2), and the total cost estimate is £1.4 billion of the total FRS budget being anticipation related expenditure.

3.8 Wider anticipation costs

FRS Partners completing HFSCs

A large number of prevention activities are carried out by FRS partners or volunteers, as opposed to paid FRS staff. The latter activities will be included in 3.7. However, there is an economic cost of volunteers and other FRS partners completing HFSCs, as this time could have otherwise been spent on alternative productive economic activity. In year ending March 2020, there were 16,356 HFSCs undertaken by FRS partners (3% of all HFSCs (HO, 2023, FIRE1201). It is assumed these visits take approximately 2 hours each (HO, 2022a), and this time is monetised using the Department for Transport value of a leisure hour, £7.79 (DfT, 2023, M2.1), to account for many of these activities being completed by volunteers. These figures are multiplied together to get a total estimate of approximately £0.3 million for the cost of FRS partners completing HFSCs.

Central government activity

There is also expenditure from central government on fire safety which should be classed as anticipation cost as it aims to prevent fires from occurring. This includes the cost of government campaign such as “Fire Kills”, the cost of Home Office staff who work on fire safety, and any other fire safety activity done outside of FRSs. This cost has not been monetised in this report due to a lack of data availability but could be considered in future work.

Referrals

FRSs carry out numerous prevention visits annually, the cost of which is captured in section 3.7. These visits can also lead to referrals, in which the FRS refer the individual to another government agency or non-government organisation. These referrals have a societal cost, as they require individuals to act, which can have opportunity and potentially financial costs. However, as these costs are not directly fire related (the referral will usually be for non-fire related reasons such as safeguarding concerns), these costs are not included in the total cost.

3.9 Total anticipation costs

Table M3 earlier in the report shows the total anticipation cost of fire. The total estimate for cost of anticipation is £8.8 billion. Some of this cost could be split into specific dwelling or other building expenditure, however it is not possible to do this for all areas of cost (including defensive expenditure on consumer goods, and FRS expenditure). As these are significant areas of expenditure, total anticipation costs by fire type are not presented. The majority of the cost is driven, unsurprisingly, by defensive expenditure in buildings; with much of the other costs being defensive expenditure in consumer goods, and FRS anticipation spend.

It is important to flag that this cost represents a large increase in anticipation costs compared to the DCLG 2011 reports (DCLG, 2011a, DCLG, 2011b). DCLG estimated that cost of anticipation was approximately £3.2 billion in 2008, equivalent to £3.8 billion in 2019 prices. The increase has been driven by the increase in building defensive expenditure costs from an increasing market size, and more importantly, the monetisation of additional areas of anticipation cost. Defensive expenditure in consumer goods, training in businesses and suppression systems were not included in the DCLG reports, and FRS spend has been re-categorised since that report, so a large proportion of spend is now classed as anticipation related, as opposed to response related. Accordingly, much of the increase has been driven by an improved ability to monetise costs, as opposed to necessarily an underlying large rise in this aspect of cost.

4. Costs as a consequence of fire

This chapter provides an overview of the methods used, and total values estimated, for the costs as a consequence of fire. These include:

• physical and emotional harms
• direct and associated property damage
• lost output
• health services
• environmental costs
• insurance administration
• wider consequences

4.1 Physical & emotional harm

4.1.1 Approach

Some individuals will suffer considerable injuries as a result of fires. These injuries will likely affect their quality of life and general wellbeing, hence the physical and emotional harm to an individual resulting from these injuries represents a cost, which can be monetised. To calculate this for different injuries, information and supporting evidence is needed to quantify the harm to an individual. The overall approach taken to quantify this cost builds on previous publications, such as the economic and social cost of crime report (HO, 2018) which have used the QALY (quality adjusted life year) approach to monetise the physical and emotional harm of an injury.

A QALY is equal to one year of life in perfect health and is used as a measure of injury or disease burden. In other words, one QALY is the value to an individual of being at full health for one year. The QALY value given by HMT (2022) is £70,000, and this is used throughout the analysis in line with DHSC best practice. The impact of an injury on a QALY primarily comes from work completed by Salomon et al., (2015) with additional studies such as GBD (2019) used for specific injuries. The other important aspect of QALY loss to consider, is the duration of the injury. Some injuries will only last for a short period of time, whereas others may impact an individual’s QALY for the whole year, or even longer.

The starting point of the calculation is the number of fire related physical injuries. Whenever an incident is responded to by a FRS, details are logged onto the IRS. Within the IRS, injuries observed at the incident by FRSs can be recorded by 4 severity categories:

1. Precautionary check.
2. First aid given.
3. Hospital admission, injuries appear to be slight.
4. Hospital admission, injuries appear to be severe.

The injuries themselves can be recorded as one of the 24 most common injuries, for example, burns, smoke inhalation, fracture, or as ‘other’, which then enables the individual inputting the data to write in a specific injury, for example, ligament injury, panic attack.

The cost of each injury is calculated as the percentage reduction in quality of life (REDUCEQL) multiplied by the duration of the injury (DUR) as a fraction of a total year, and by the value of a QALY, to give an estimate of the average cost associated with the fire related injury. This is done for each injury type.

The formula, and a worked example, are presented below:

REDUCEQL * DUR * QALY = Average physical cost

Serious hospitalising burns have a QALY impact (REDUCEQL) of 45.5%, and an average duration of 182.5 days (0.5 years).

Multiplying these 2 figures by the annual QALY (£70,000) gives the injury cost of £15,925 (0.455 * 0.5 * £70,000).

The cost of fatalities was also calculated in this way, with the duration being expected years of life left. This is further explored in section 4.1.7.

The QALY approach is the best available method of estimating the cost of fire-related injuries, however the QALY used is an average across the population, and may differ depending on an individuals’ characteristics, for example, age.

4.1.2 Number of fire-related physical harms

Using the IRS, we were able to count the number of physical injuries sustained from fire incidents. This data is directly inputted by the responding FRS of each incident, detailing variables on incident nature and victims. Raw data was used, as opposed to estimates for injury severity. Annex table A3 shows the number of physical injuries, by severity, recorded in the IRS for year ending March 2020.

However, when recording an incident, FRSs only include the main injury they feel the individual suffered, ignoring potential secondary injuries that also lead to physical harm. To account for these not being recorded, the initial recorded injury was analysed to determine potential associated secondary injuries. For example, an individual with severe burns will also likely have a degree of smoke inhalation. This injury was assumed to be one level of severity below the main recorded injury. For example a burns hospital admission with “injuries that appear to be severe”, would be accompanied by overcome by fumes hospital admission which “appears slight”, and a back/neck injury would be accompanied with some bruising. All primary injuries were examined to identify any secondary injuries, however not all primary injuries were expected to have secondary injuries. Annex table A4 shows the primary injuries and their associated likely secondary injury.

4.1.3 Quality-adjusted life loss associated with physical harms

To calculate the percentage reduction in a person’s quality of life (REDUCEQL) from different injuries, disability weights were used. The disability weights come from Salomon et al., (2015) and GBD (2019). Where the injuries in these papers do not exactly match the injuries suffered as a result of fire (from IRS data), the best proxy is taken. Central QALY impact values from the 2 papers are taken throughout. There is a natural level of uncertainty in these estimates, because all injuries will have slightly different impacts on individuals, and confidence levels are provided in both reports used. We have followed the approach of previous Home Office reports in only estimating a central cost of fire. As an example, the impact on an individual’s QALY from burns is expected to vary from 1.6% to 45.5% depending on severity (with precautionary checks having the lowest figures, and serious hospital admissions having the highest). The same figures for back/neck injuries, is 1.6% to 31.2%. Annex table A5 presents the full REDUCEQL data for each injury type.

4.1.4 Duration of physical harms

Harm durations (in days) were taken from Dolan et al., (2007) with additional sources used for specific injuries and best proxies used where little literature was available. Annex table A6 shows the full breakdown of durations for every injury type. As an example, burns injuries were assumed to last between 2 days and 182.5 days (half a year) depending on severity, and back/neck injuries were assumed to last between 1.5 and 90 days.

The duration assumptions used are the best available estimates for the average length of harm. However, as some fire related injuries have long term impacts, long-term harms were also considered within the “victim went to hospital, injuries appear to be serious” category of injury, for harms that may extend beyond the duration estimates for these injuries in annex table A6. The exact nature of long-term harms is highly uncertain, as they will likely vary by injury type, and the circumstances surrounding the injury. Future work could look at linking medical and IRS data, to get an actual estimate for harm, however this was not possible in this report, and so assumptions were made.

There is precedent in the academic literature that serious burns injuries (used as a broad proxy for fire-related injuries) have long-term impacts on individuals (James et al., 2020) that can persist for multiple years. There are limited long-term follow up studies, although there is evidence that the quality of life of burns victims can remain below the general population for multiple years after the injury.

To cite available literature, (Choinière et al., (1991)) completed a follow-up survey a year after a burn injury, and found that a significant proportion of individuals still had pain, and that their burns impacted their work and social lives. Similar results, over longer time periods were found by Öster et al., (2011), who found that 2 to 7 years after an injury pain and discomfort levels were above the general population and Dauber et al., (2002) who found that 12 years after serious burn injuries individuals still suffered with pain and had their injuries impact their daily lives. Moi et al., (2016), and Holavanahalli et al., (2010) completed even longer term studies (16 and 17 years after injury respectively) and found significant burns related issues persisted. These studies provide strong evidence that this cost should be included in the report.

With a lack of better data, the 12 years of impact from Dauber et al., (2002) is taken as the best estimate for length of harm from serious fire injuries, and it is assumed that 50% of serious injuries would lead to these long-term harms. The 50% is an assumption made with a lack of better data, and aligns with the rounded average impact from academic research which quantifies long-term impact (Choinière et al., (1991) find that 35% remain in pain, and 67% have their injury impact their social life, and Dauber et al., (2002) find that 52% have long-term pain, 55% have their injury impact their daily life).

This 12-year period will be referred to as ‘long-term’ throughout the paper, and the 50% assumption is used whenever long-term harms are quantified^{[footnote 24]}. The QALY impact of the respective injury was assumed to fall at a consistent rate across the 12 years (to 0 the year after) to account for evidence that there would be improvements in physical health over the period (Moi et al., 2016). This rate of improvement will vary be individual, and for some individuals will likely be non-linear, or not reach zero.

A discount rate of 1.5% was also applied annually over the period (HMT, 2022) to account for social time preference, defined as the value society attaches to present, as opposed to future, consumption. This discount rate is lower than the standard annual discount rate of 3.5%. The use of this lower discount rate for health outcomes is recommended in the Green Book and reflects the absence of a ‘wealth effect’ in the discounting of health outcomes (HMT, 2022).

4.1.5 Costs of physical harms

Using the duration, QALY impact and QALY values outlined above, and then considering long-term and secondary harms, the unit cost per injury type could be calculated. Annex table A7 provides a full breakdown of the unit cost per injury, per severity. The unit cost of a burns injury varies from £6 if just a precautionary check occurs, to £226,216 for serious hospital admissions.

The unit costs of injury were then multiplied by the number of incidents of each fire type to calculate total costs. The total costs by fire type and accidental/deliberate fires are presented in table M9 below, and these sum up to all physical harms suffered from fires in year ending March 2020. Unit costs by fire type are also shown (calculated as total cost divided by number of incidents). The unit costs are presented as a function of all fires attended, just primary fires attended, and of just fires attended that led to a physical injury.

Table M9: Cost of physical harms by fire type

Fire type	Total cost	Total injury incidents	Unit cost (Total incidents)	Unit cost (Primary incidents)	Unit cost (Injury incidents)
Other Buildings	£5.9m	879	£370	£410	£6,740
Dwellings	£23.7m	5,155	£750	£830	£4,590
Other Outdoors	£6.4m	352	£80	£1,190	£18,230
Road Vehicles	£3.3m	551	£160	£160	£5,980
Total	£39.3m	6,937	£260	£570	£5,670
Deliberate	£7.5m	1,025	£110	£390	£7,320
Accidental	£31.8m	5,912	£380	£640	£5,380

Table M9 notes:

1. Unit costs rounded to the nearest £10, and total costs to the nearest £0.1 million.
2. For the purpose of the total incidents attended figure, secondary fires are included in an overall IRS category, for instance, chimney fires are included in dwellings, and secondary outdoors are “other outdoors”.

4.1.6 Costs of emotional harms

There is growing academic evidence to suggest that individuals can also suffer emotional trauma and harm as result of fires. These impact both those who are physically injured, and those who witness the event. However, this section only considers emotional harms to individuals who are also physically injured.

The likelihood of an individual suffering emotional harms is proxied using ONS data (ONS, 2020c) on the likelihood of emotional harm following an arson offence, with a 50% reduction applied for injuries that do not lead to a hospital admission. Arson is deemed to be an appropriate proxy due to the traumatic nature of both arson and serious fire injuries. Estimations for QALY impact (REDUCEQL) and duration (DUR) are taken from the 2018 cost of crime report (HO, 2018 and Salomon et al., 2015), with the duration of fear (lasting over a year) being discounted. These are both kept constant across injury severities, with only likelihood of emotional harm being changed. The formula for emotional harms is as follows:

LIKELIHOOD * REDUCEQL * DUR * QALY (£70,000) = Average emotional cost

Table M10 shows the unit cost of the different monetised emotional harms by their severity. They were found using the above equation for three emotional harm categories: fear, depression, and anxiety/panic attacks, and then summing the results.

Table M10: Unit costs of emotional harm

Harm type	Likelihood	Duration (years)	REDUCEQL	Cost
Fear
Non-hospital	13%	1.25	3%	£340
Hospital	26%	1.25	3%	£680
Depression
Non-hospital	11%	0.17	15%	£178
Hospital	21%	0.17	15%	£356
Anxiety/Panic Attack
Non-hospital	11%	0.17	13%	£163
Hospital	21%	0.17	13%	£327

Table M10 notes:

1. Likelihood varies depending on if the incident resulted in a hospital admission or not. Incidents resulting in hospitalisation are more likely to cause emotional harms. 0.17 years equivalent to approximately 2 months, and 1.25 years to 15 months.

The unit costs were then multiplied by the number of injury incidents with that severity to give an overall cost. Table M11 presents the totals by incident type and split into accidental and deliberate fires.

Table M11: Cost of emotional harms by fire type

Fire type	Total cost	Total injury incidents	Unit cost (Total incidents)	Unit cost (Primary incidents)	Unit cost (Injury incidents)
Other Buildings	£0.8m	879	£50	£60	£920
Dwellings	£5.0m	5,155	£160	£180	£980
Other Outdoors	£0.4m	352	£0	£70	£1,050
Road Vehicles	£0.6m	551	£30	£30	£1,010
All fire types	£6.8m	6,937	£40	£100	£980
Deliberate	£1.0m	1,025	£10	£50	£1,010
Accidental	£5.7m	5,912	£70	£120	£970

Table M11 notes:

1. Total costs rounded to the nearest £0.1 million, and unit costs are rounded to the nearest £10.

4.1.7 Cost of Fatalities

The cost of harm from fatalities were calculated by multiplying the expected years left of life (ONS, 2021c), by the QALY and applying a 1.5% annual discount rate (HMT, 2022). The IRS records data on the sex and approximate age of a victim at the time of death, which could be used to estimate expected years left of life using ONS life expectancy data. The IRS data records the age of a victim within a range, and the mid-point of this range was taken. The central cost estimate for all fire types was £314 million, shown in table M12, split by fire type.

Table M12: Cost of fatalities by fire type

Fire type	Total cost	Total fatality incidents	Unit cost (Total incidents)	Unit cost (Primary incidents)	Unit cost (Fatality incidents)
Other Buildings	£27.5m	18	£1,700	£1,900	£1.5m
Dwellings	£245.5m	200	£7,800	£8,600	£1.2m
Other Outdoors	£18.4m	11	£200	£3,400	£1.7m
Road Vehicles	£22.8m	16	£1,100	£1,100	£1.4m
All fire types	£314.2m	245	£2,000	£4,600	£1.3m
Deliberate	£91.6m	58	£1,300	£4,800	£1.6m
Accidental	£222.6m	187	£2,600	£4,500	£1.2m

Table M12 notes:

1. Total costs rounded to the nearest £0.1 million, and unit costs rounded to the nearest £100, or £0.1 million.

4.1.8 Cost of rescues

Needing to be rescued from a fire related incident may also lead to emotional trauma even when no physical injury occurs. The impact of this harm was costed using the same methodology as the emotional harms section (4.1.6)

It is assumed that being rescued from a fire is as emotionally harmful as a non-hospital physical injury, and that the likelihood of experiencing emotional harm is the same. There is limited evidence in this area, so this was seen as the most appropriate proxy. Therefore, the unit cost of the emotional impact from rescues totals £681 (see table M10, £340 + £178 + £163). This was then multiplied by the number of people rescued to get a total cost estimate in year ending March 2020, which was 3,015, to give a total cost estimate of £2.0 million.

4.1.9 Summary

Table M13 shows the total costs of all physical and emotional harms by fire type, with a total overall cost and per incident unit cost.

Table M13: Total physical and emotional harm costs

Fire type	Total cost	Total fire incidents	Total primary fire incidents	Unit cost (Total incidents)	Unit cost (Primary incidents)
Other Buildings	£34.6m	16,042	14,328	£2,200	£2,400
Dwellings	£275.8m	31,651	28,491	£8,700	£9,700
Other Outdoors	£25.2m	85,202	5,386	£300	£4,700
Road Vehicles	£26.7m	21,214	20,560	£1,300	£1,300
All fire types	£362.3m	154,109	68,765	£2,400	£5,300
Deliberate	£100.7m	69,850	19,159	£1,400	£5,300
Accidental	£261.6m	84,259	49,606	£3,100	£5,300

Table M13 notes:

1. Total cost rounded to the nearest £0.1 million and unit costs rounded to the nearest £100.

4.2 Direct and wider property damage

4.2.1 Direct property damage (rebuild)

Fires can lead to significant damage to property, and the impact of this needs to be monetised and included in this report. To calculate this cost, 2 data points are required:

1. The area of property damaged (in m²)

2. The unit cost of property damage (per m²)

The first comes from IRS data on approximate property damage in m² per incident. In the IRS, this data is collected for every fire incident, and comes within a banded range (for example, 0, Up to 5, 6 to 10, 11 to 20). Throughout this analysis, the midpoint of the range is taken for each incident as it is assumed that property damage is distributed equally within each band. A low and high range is presented in section 4.2.3 which assesses how the costs would change if the low or high figure in each band were used instead. Where property damage is recorded as being “over 1,000m²”, a figure of 1,000 is used as a conservative best assumption to prevent overestimating the costs. This however means it is possible that some of the costs could be underestimations. There were only 198 incidents in year ending March 2020 that had over 1,000m² damage, out of 35,550 incidents with any property damage (0.6%), so any impact on totals is expected to be small.

The second input cost (rebuild cost per m²) comes from Building Cost Information Service (BCIS) rebuild data (sourced via Nimblefins, 2022). The range of costs are estimated to be £1,380 to £1,910 nationally in year ending March 2020 prices. These values were calculated from the average 2021 prices in the Nimblefins data and include a 15% reduction to year ending March 2021 prices (in line with the Nimblefins estimate of a 17% yearly rise between 2020 and 2021). As this then gives costs in 2020 prices (it is not specific when in 2020, however, it is likely late 2020 due to the article being published in early 2022), a further 1.4% reduction (ONS, 2022g, “Regular maintenance and repair of the dwelling” used as a proxy) is applied so the values are in mid-2019 prices. For the purposes of this report, this is assumed to be equivalent to year ending March 2020 prices.

This is a significant reduction and has been completed to ensure consistency with the rest of the report. However, it means that it is likely that this cost will rise significantly in future years due to increases in rebuild costs, even if the amount of damage stays the same. Rebuild costs are relatively uncertain and will vary by property type, area, and exact nature of damage. These rebuild costs are the best available and have been corroborated by a number of rebuild cost estimation databases. However, they are averages across all property types and exact damage costs will vary on a case-by-case basis.

Calculations were completed across 9 regions of England to account for regional differences, and the results of these calculations are displayed in table M14. IRS data is available by FRS, allowing for matching between FRS and regional location.

Table M14: Cost per m² by region (year ending March 2020 prices)

Area	Cost per m2
North-East	£1,439
Yorkshire & the Humber	£1,439
West Midlands	£1,497
East of England	£1,497
East Midlands	£1,497
North-West	£1,517
South-West	£1,575
South-East	£1,776
London	£1,990

The 2 incident types in the IRS looked at were ‘dwellings’ and ‘other buildings’, as these are the only ones with recorded property damage. ONS property price data was considered as an alternative source for property cost per m², however, these estimates likely overestimate the actual cost of damage and rebuild following a fire. Property prices are determined by several other factors such as local demand and the price of the land a property is on, not just the cost to rebuild it, so specific rebuild costs are deemed more accurate and are used in this report.

The IRS collects both fire damage and total damage, with the latter comprising the former, plus estimates for smoke and water damage. To calculate overall damage cost, recorded fire damage was multiplied by the rebuild unit costs, with smoke and water damage multiplied by 20% of the rebuild unit costs. This was done as smoke and water damage are unlikely to require a full rebuild, so a lower unit cost was used. 20% was used as the downrate factor because 20% is the average of the high and low scenario for smoke damage – relative to fire damage – in a CEBR (2014) paper. The following equations are used:

Fire damage area * rebuild cost = Fire damage cost

(Total damage area – fire damage area ) * (rebuild cost * 0.2) = Smoke/water damage cost

The total cost of direct property damage from fires in the year year ending March 2020 was calculated to be £1,051 million of which £899 million is from fire damage, and £152 million is water and smoke damage.

The costs were also calculated by property type, presented in table M15 below.

Table M15: Direct property damage totals by fire type

Fire type	Total cost	Total fire incidents with damage	Unit cost (Total incidents)	Unit cost (Primary incidents)	Unit cost (Damage incidents)
Other Buildings	£652m	12,415	£40,600	£45,500	£52,500
Dwellings	£399m	23,135	£12,600	£14,000	£17,300
All fire types	£1,051m	35,550	£22,000	£24,500	£29,600
Deliberate	£257m	6,280	£31,000	£37,500	£40,900
Accidental	£794m	29,270	£20,100	£22,100	£27,100

Table M15 notes:

1. Total costs rounded to the nearest £1 million and unit costs rounded to the nearest £100.
2. ‘All fire types’ does not cover the fire types of Other Outdoors and Road Vehicles fires.

4.2.2 Wider property damage

Damage from fires in properties do not only lead to direct rebuild costs. There are also wider impacts on society of these fires occurring, such as damage to contents and stock, business interruption and a loss of machinery. To calculate the size of this wider cost, an uprate factor was applied to the direct damage figure. Dwellings had an uprate factor of 0.28, whilst other buildings had a factor of 1.33 to account for the greater potential for costs in a business setting^{[footnote 25]}. In simple terms, this equates to 28% or 133% additional cost being added. If a dwelling fire had £100 of damage, the new estimate would be £128, and if an “other building” fire had £100 of damage, the new estimate would be £233. These factors were found using a Fire Protection Association (FPA, 2020) analysis of costs in warehouse fires. Although these estimates apply to warehouses, these are seen to be the best overall estimations for the size of the cost of wider property damage, and table M16 demonstrates how these cost estimates are converted into an uprate factor that can then be used for dwelling fires.

Table M16: Uprate factor for “other buildings” and dwelling fires

Cost type	Proportion of total damage cost, according to FPA analysis	Other buildings	Dwelling
Property damage	43%	YES	YES
Business interruption	18%	YES	NO
Damage to contents	7%	YES	YES
Loss of machinery	5%	YES	NO
Damage to stock	5%	YES	YES
Loss of rent	2%	YES	NO
Other	20%	YES	NO
	Sum of in-scope proportion	57%	12%
	Uprate factor from property damage	1.33	0.28

Table M16 notes:

1. Uprate factor calculated through summing all relevant cost areas (marked YES) and then dividing by the 43% property damage figure. So, business uprate is 57% (18 + 7 + 5 + 5 + 20)/43%, and dwelling uprate is 12% (7 + 5)/43%.

Wider damage costs are estimated by multiplying the damage cost of each incident by the relevant uprate factor, which gives £959 million. This is then summed with direct property damage to give a total property damage cost estimate of £2,009 million. Table M17 presents the total damage cost by fire type, and annex table A8 presents the total wider property damage cost by fire type. Property damage is by far the largest consequence cost to society from fires.

Table M17: Total damage costs by fire type

Fire type	Total cost	Total fire incidents with damage	Unit cost (Total incidents)	Unit cost (Primary incidents)	Unit cost (Damage incidents)
Other Buildings	£1,499m	12,415	£93,400	£104,600	£120,700
Dwellings	£511m	23,135	£16,100	£17,900	£22,100
All fire types	£2,009m	35,550	£42,100	£46,900	£56,500
Deliberate	£517m	6,280	£62,300	£75,400	£82,300
Accidental	£1,493m	29,270	£37,900	£41,500	£51,000

Table M17 notes:

1. Total costs rounded to the nearest £1 million and unit costs rounded to the nearest £100.
2. ‘All fire types’ does not cover the fire types of ‘other outdoors’ and ‘road vehicles’ fires.

4.2.3 Evaluation of property damage method

This is the largest cost area in the report, so an evaluation of the method was undertaken to assess how sensitive the total numbers were to alternative assumptions.

The process was also conducted using property value per m² (values accounted for regional differences) (ONS, 2021d) instead of rebuild cost

The result was significantly higher in total than rebuild cost. It was decided that this approach was not as suited to the aim of the report, as incidents do not always destroy all value of the associated property. In addition, house prices do not equal rebuild costs and are more susceptible to the forces of supply and demand, so are likely to be a less accurate estimates for the true cost of damage.

Many of the fields in the IRS, including those on fire spread that are used as an input in this section, are completed by firefighters at the scene using their best estimates for the impact of the incident

It can be hard to estimate metre squared damage from looking at a fire, meaning the data has some inherent uncertainty in it. This must be considered when using the results of this report and means there could be some over or under estimation. However, it remains the best available source for fire and rescue incident data.

To evaluate the potential range in total costs, the low and high figures for each damage estimate band were used instead of the mid-point

For example, for the 6 to 10m² band, 6 and 10 will be used in the low and high scenarios instead of 8 (as per the central estimate). Incidents recorded as 1000m² are kept constant at 1000 throughout scenarios as there is no range for this band and as previously stated, these are only a small number of incidents so the likely impact on the totals is small. Using the low or high figures within each band alters the central estimate by approximately 33% for both the low (£1.36 billion) and high (£2.66 billion) estimate, showing this to be a sensitive cost area within this report.

The Association of British Insurers (ABI) collect data on the amount paid out by insurers as a result of fires. In 2018, £1.3 billion was paid out to customers by insurers (ABI, n.d.)

Uprating this to year ending March 2020 figures, and accounting for the ABI not covering the whole insurance market, this estimate is approximately £1.7 billion in year ending March 2020. This is considered to be broadly consistent with the estimates calculated in this report, when considering the differences in methods. Total insurance pay-outs are expected to be lower than the total property damage estimates for several reasons. Firstly, they only include insured properties, so do not capture all damage. Secondly, they are mainly related to fire damage, so may not fully consider smoke and water damage. Thirdly, although they do consider business interruption, they likely do not include the full range of wider damage impacts outlined in annex table A8.

Total property damage splits the cost into fire damage, and then smoke and water damage. It is expected that the latter will not have the same impact as fire damage, and so will cause less damage in impacted areas

There is limited evidence on the difference between the cost of repairing fire damage compared to water and smoke damage. However, we expect that this rebuild cost will be lower. In the absence of better data, it is assumed that the rebuild cost of water and smoke damage is approximately 20% of the overall rebuild cost. This is highly uncertain but considered to be the most suitable estimate with the data available. This gives a total direct property damage cost of £1.05 billion. The impact of this assumption is tested. If it were to be 25%, then the cost would be £1.08 billion, whereas a 50% assumption gives a total of £1.27 billion. The equivalent range, including wider damage, is £2.1 billion to £2.4 billion, with an estimate of the £2.0 billion presented in the report.

4.2.4 Road vehicle damage

So far in this section, only the damage to properties because of fires has been costed. However, there can also be significant damage to road vehicles because of fires. The IRS records data on all road vehicle fires, which state the vehicle type (for example, car, motorcycle) and the area of damage from the fire (for example, fuel tank, engine compartment). There are 14 vehicle types and 8 damage categories in total.

The first step was to find the average cost of each of the 14 vehicle types. This was done using used vehicle cost prices as all damaged vehicles would be used, not brand new. Data was collected from a range of online data sources for these calculations, with proxies used where data was uncertain. An average is used, however there remains a relatively high level of uncertainty in this estimate as there can be big variations in the repair cost of vehicles following a fire. In addition, there can be a large variation in the actual vehicle, within the vehicle type classifications in the IRS. Different cars can have largely different repair costs based on their manufacturer, or specification (for example, electric versus petrol cars). The average cost per road vehicle, and number of in-scope incidents used, are outlined in annex table A9.

Once the cost of the whole vehicle was calculated, the damage categories were used to estimate the cost of each incident. The damage categories were each given a percentage of expected damage to the vehicle (from 100% for “whole vehicle”, to 25% for “roof/roof rack”). These were then multiplied by the estimated cost of the whole vehicle to calculate costs. As a worked example, an incident involving a car in which the fire was in the engine compartment, would be costed as £14,029 x 60% = £8,417.

This was conducted for all incidents, giving a total cost of £272 million. Table M18 gives the total and unit costs, by fire type, for the damage to road vehicles from fire in year ending March 2020.

Table M18: Total and unit costs of road vehicle damage

Fire type	Total cost	Total fire incidents with damage	Unit cost (Total incidents)	Unit cost (Primary incidents)	Unit cost (Damage incidents)
Road Vehicles	£272m	20,252	£12,800	£13,200	£13,400
Deliberate	£116m	9,729	£11,300	£11,800	£12,000
Accidental	£155m	10,523	£14,200	£14,500	£14,800

Table M18 notes:

1. Total costs rounded to the nearest £1 million and unit costs rounded to the nearest £100.
2. Figures may not sum due to rounding.

The cost of both road vehicle and property damage only considers the cost of the damage to each of these directly (the property or the vehicle). The cost does not include any potential environmental costs to areas around the impacted area (for example, the ground next to a vehicle, or outside a property). There is no data to estimate the size of the additional impact, so it has not been quantified or monetised. However, the environmental costs section of this report does consider the impact of emissions from these fires, and the cost of general outdoor fires on the environment.

4.3 Lost output

4.3.1 Approach

A reduction in output due to fire related incidents can lead to a large societal cost. For example, if a high street store is damaged, then individuals working in the store may be unable to work or be made unemployed. This can lead to a reduction in output. Similarly, if a worker is injured, be that physically or emotionally, they may be less productive at work or absent entirely whilst they recover. In this section, both these cost areas are estimated.

4.3.2 Lost output resulting from property damage

Property damage to buildings may prevent individuals being able to work or render them unemployed, leading to lost output. This cost differs from the business interruption cost monetised in section 4.2, as it is personal losses of output instead of total business losses, and from a methodological point of view, insured losses of output which do not lead to unemployment would be covered in wider property damage, whereas uninsured costs which lead to a loss of employment would be included here.

Due to the potential complexity of this cost area and the data available to us, it was decided to explore the effect of property damage on output through lost individual employment. It was thought to be a reasonable equivalency as it is assumed that individual salaries represent their marginal product of labour (output for the employer). The analysis used in this section is based on a CEBR paper (CEBR, 2014)^{[footnote 26]} which found that there are 384 jobs directly lost in the warehouses assessed. By working backwards through their calculations of total damage area and unemployed numbers, it was estimated that for every 605m² of damage in warehouses, an individual lost employment.

This was used as a proxy for individuals who are made unemployed, or are simply unable to work, because of property damage from fires. To assess the impact of property damage on wider economic activity across all industries, IRS data on fires in seven different commercial industries^{[footnote 27]} was used. Total damage was used instead of just fire damage because it gives a more accurate picture of the full impact on businesses from a fire, given that water and smoke damage can be very impactful.

The CEBR paper only considered warehouse fires, so to calculate the total cost for different industries, adjustments were made to account for differences in employment density, average unemployment durations, and salaries. These considered labour market flexibility and ability to work from home too.

4.3.2.1 Employment densities

Employment densities were considered because the same size fire will have a much bigger impact on workplaces that are more densely populated. A fire that could destroy a retail premises, may have a much smaller impact in a warehouse because of the difference in size of premise and employment density. Densities were found using analysis by the Homes and Communities Agency (2015), which gave a corresponding m² figure for an employee in the given industry. We were then able to find and compare the densities of different industries to warehousing. These densities were then downrated using ONS (2022h) data on the proportion of hybrid working available by industry, to account for the ability for some individuals to work from home. Warehousing was used as the baseline, with other industries then adjusted by how proportionally different they are to this sector (which fits in the transportation and storage SIC code). This is most easily explained in a worked example:

Retail premises are assumed to have a workplace density of 17.5.

In 2020, 4.7% of warehousing businesses using home-working, whereas 7.3% of those in the wholesale and retail trade could. So, the density is multiplied by 1.6 (7.3 divided by 4.7) to give an adjusted density of 27.2.

Average damage in each retail fire could then be divided by 27.2, to give number of individuals who lost employment as a result of it.

The density of hotels, boarding houses and hostels was not included in the Homes and Communities Agency (2015) report, so alternative data (PWC, 2018; ONS, 2013; ONS, 2021f) were used; on average, employee numbers, per bed place, hotel numbers, and site sizes (accounting for open areas) to get to a density figure. Table A10 in the annex presents the employment density and conversion factor for all 7 industries.

There are important caveats to consider in this analysis. Firstly, although there is some consideration of the ability for individuals to work from home in year ending March 2020, this does not consider the impact that COVID-19 has had on individuals in some industries that are now working more frequently from home. This could reduce this cost in future as some industries and premise types (such as offices) may be more resilient to fires in the workplace as employees could work from home with more limited impacts on productivity. Secondly, this analysis assumes that volume of damage is the only factor to consider in estimating the impact on a business. The location of a fire is also crucial. A fire could be smaller but impact key machinery or a key element of a premises, which would make it more impactful on businesses. It is not possible to work out the location of every fire, so this is a natural caveat to the analysis. Finally, the analysis is based on warehousing as a proxy. This assumption is the best possible with the available data, but there may be differences between service and goods-related jobs in terms of the impact fires have on employment, beyond what is included in our adjustments.

4.3.2.2 Unemployment durations

Unemployment durations for individuals impacted by warehouse fires were also included in the CEBR paper. These were broken down into 4 timeframes (under 6 months, 6 to 12 months, 12 to 24 months, and over 24 months), with the proportion of staff unemployed for that duration of time also given (46%, 18%, 19%, 17%). In this analysis, mid-points were taken between each range, with 24 months used as the highest estimate.

The proportion of staff unemployed was adjusted for other industries using ONS data (2022b) on unemployment duration by industry to account for varying levels of labour market flexibility. This data gave a proportion of those unemployed for over 12 months and below 12 months, which was then compared to ONS data on warehousing in order to adjust for the industry. A worked example is presented below:

CEBR data showed that 36.1% of those previously employed in the warehousing industry were out of work for over one year, whereas this was 17% for the retail industry according to ONS data. This is 48% as much.

The CEBR proportions over one year are therefore multiplied by 0.48, and the proportions below one year are uprated to account for the remaining 83%.

This gave the following data:

Unemployment duration	Warehousing	Retail
Under 6 months	46%	59%
6 to 12 months	18%	23%
12 to 24 months	19%	9%
Over 24 months	17%	8%

4.3.2.3 Costs

To calculate the cost of each fire that led to property damage, we took the median annual salary by industry (ONS, 2021e) and multiplied it by unemployment durations and the number of individuals expected to be impacted by the fire. This method assumes that salary is equal to marginal product of labour. It was not possible to monetise returns to capital or producer surplus due to a lack of available data. As there are 4 unemployment duration timeframes the calculation was repeated 4 times for each industry.

There are some fire incidents that do not lead to enough property damage to meet the threshold for an unemployed employee as per the employment densities calculations. For example, a retail fire with 30m² of retail damage does not hit the 27.2 m² average density in section 4.3.2.1. These were excluded from the calculations to prevent overestimating the cost. The costs which lead to an impact over a year in the future are discounted appropriately at 3.5%. A worked example of the calculations is presented below:

A fire in a food and drink premise resulted in 35.5m² of damage. The damage area to render one person unemployed for this industry is 17.1². Therefore 2.1 persons are assumed to be impacted.

All individuals unemployed from food and drink premises fires are summed, giving 2,278 individuals, and the median salary is £22,600.

The unemployment duration adjustment is then applied, giving 61%, 24%, 8% and 7% respectively.

Therefore, the equations are:

2,278 * (£22,600 * 0.25 * 61%) = £9.2 million

2,278 * (£22,600 * 0.75 * 24%) = £10.9 million

2,278 * (£22,600 * 1.5 * 8%) * 0.99 = £7.3 million

2,278 * (£22,600 * 2 * 7%) * 0.98 = £8.8 million

Total cost: £36.1 million

The total cost across all in-scope industries is calculated to be £118.6 million. The total and unit costs, by fire type, of lost output resulting from property damage is shown in table M19.

Table M19: Total and unit costs of lost output resulting from property damage

Fire type	Total cost	Total individuals impacted	Unit cost (Total incidents)	Unit cost (Primary incidents)	Unit cost (individuals impacted)
Other Buildings	£119m	6,200	£7,400	£8,300	£19,100
Deliberate	£30m	1,600	£5,800	£7,600	£18,600
Accidental	£89m	4,600	£8,100	£8,500	£19,300

Table M19 notes:

1. Total costs rounded to the nearest £1 million and unit costs rounded to the nearest £100.
2. Total individuals impacted refers to the number of individuals who have lost employment as a result of property damage.

4.3.3 Lost output (from injury)

Approach

This section estimates the cost of lost productivity resulting from individuals being victims of fire related incidents. Victims may take time off work due to the incident and may also be less productive at work for some time after it. The following costs are included:

(i) Time taken off work resulting from the fire related incident.

(ii) Reduced productivity at work as a result of physical and emotional injuries sustained from the fire related incident.

We assumed that those suffering with physical and emotional injuries would be less productive at work for the duration of the injury. The QALY loss associated with the injury is used as a proxy for the extent of their reduction in productivity. The QALY losses used to estimate the physical and emotional costs are assumed not to already capture lost productivity, and so the cost of QALY loss described in section 4.1.1 is separate to the cost of lost output from injury described in this section. This is supported by other Home Office reports (HO, 2018), which conclude that any double counting between QALY and productivity loss is negligible. We therefore assume we are not double counting lost productivity by including an estimate for it in this section.

(iii) Lost output resulting from fatalities.

This multiplied yearly average salaries (including non-wage costs) by expected years before retirement.

Time off work

Many individuals are likely to take time off work following a fire related injury. To calculate this time off, we used the four severity categories within the IRS as opposed to specific injury types. We completed a literature review on return to work times by injury severity. For precautionary checks and first aid injuries, it was assumed that each individual would take off approximately 5.5 hours (HO, 2018). For those requiring hospital treatment, it was assumed that for slight hospital admissions, individuals will take off approximately five weeks, and for severe hospital admissions, they would take off 17 weeks (Esselman et al., 2007).

The number of injuries that occurred to individuals of working age was then multiplied by the average year ending March 2020 employment rate of 76.2% (ONS, 2022a), the number of hours off work and the average hourly employment cost of £16.16 (ONS, 2022c) (including an 18% uprate for non-wage payments) to get the overall cost for each of the four severities. The total cost was £6.7 million, which is set out by fire type in annex table A11.

Reduced productivity at work following an injury

Reduced productivity arises because those who have suffered physical and emotional harms from fire related incidents are likely to be less productive at work. For example, someone who has burned their hand may find it difficult to use a computer keyboard and therefore may be less productive than their usual level. This concept was first monetised in the economic and social cost of crime report (HO, 2018). Reduced productivity is estimated by first calculating the average number of hours of productive labour lost due to physical and emotional harms. This is calculated by multiplying the duration of harm, expected productivity loss, and average hours worked (ONS, 2022d). The former two match the assumptions in section 4.1.

This figure is combined with the number of injuries that occurred to individuals of working age to calculate the total number of productive hours lost. This is then multiplied by average employment costs and the employment rate to give a total estimate for the average cost of reduced productivity per injury type (ONS, 2022a and 2022c). This method assumes that salary is equal to marginal product of labour. It was not possible to monetise returns to capital or producer surplus due to a lack of available data.

This is broadly the same approach as physical harms. However, instead of a QALY value and the duration of a year, the average hourly employment costs and average hours worked a year were used, with employment rates considered. Long-term impacts are considered, but secondary impacts are not, and all figures are discounted at 3.5%.

Emotional costs are also factored in. These use the exact same approach as the emotional harms QALY section, except they make the same adjustments as the physical reduced productivity section in that they account for employment rates and average salaries, instead of QALY values.

This cost only includes the direct impact on adults. There is also likely to be a cost to children or family carers who are required to care for their parents who are injured in a fire. This has an opportunity cost as this time would either be spent in education (in the case of children), or in undertaking other productive activities (for other carers). This impact, both in terms of direct cost and long-term productivity, is not monetised due to a lack of data.

Annex table A12 shows the total and unit cost of reduced productivity, by fire type, caused by the physical and emotional harms from fire. The total cost of reduced productivity (both physical and emotional) is estimated to be £9.6 million.

Fatalities

The lost output from fatalities was calculated by estimating the years left until retirement and using average hourly wage (plus non-wage costs) as a proxy for hourly output. With the assumption that each individual who suffered a fatality would have retired at 65, those who were over 65 years old at the time of the fatality were excluded from the calculation. The following equation was used for each fatality, and this was summed across the 133 in-scope fatalities:

Years until retirement x Hours worked per year x Average hourly employment cost

This equation only calculates the lost output cost of each fatality, and so it is separate (and in addition) to the QALY loss discussed in section 4.1.7. Given that IRS data gives the age band of each fatality, the years until retirement of each fatality was calculated by subtracting the retirement age (65) by the midpoint of the age band of the fatality. The total hours of work lost as a result of a fatality is also downrated by the average monthly employment rate in the year ending March 2020 for 16 to 64 year olds, taken from ONS (2022a) data. Annex table A13 shows the total and unit costs, by fire type, from the lost output from fatalities in year ending March 2020. The total cost of lost output from fatalities is estimated to be £35 million.

4.3.4 Summary

Annex table A14 gives a breakdown of the cost of lost output from fire by type of lost output, and table M20 give a breakdown of the cost of lost output from fire by fire type. Total lost output is estimated to be £170 million.

Table M20: Total lost output costs by fire type

Fire type	Total cost	Unit cost (Total incidents)	Unit cost (Primary incidents)
Other Buildings	£125m	£7,770	£8,700
Dwellings	£36m	£1,150	£1,280
Other Outdoors	£5m	£60	£890
Road Vehicles	£4m	£200	£210
Total	£170m	£1,100	£2,480
Deliberate	£45m	£640	£2,340
Accidental	£125m	£1,490	£2,530

Table M20 notes:

1. Total costs rounded to the nearest £1 million, and unit costs rounded to the nearest £10.

4.4 Health Services

4.4.1 Approach

Health service costs aim to capture the costs to the NHS and other healthcare providers as a result of the physical and emotional harms of fire. If fewer people were injured in fires, the resources used to treat them could be used in alternative productive activities; meaning costs to health services resulting from fire, are a social cost. The estimates of health service costs are based on assumptions regarding the treatment likely to be required for certain injuries based on their severity. Only incidents recorded in the IRS leading to a hospital admission are included in the calculations, as it not expected that less severe injuries will incur any health service costs. Data on injury severity is based on firefighter assessments and simply whether the individual went to hospital. It is not certain how long any individual remained in hospital after the injury, or how long they received treatment for, so there may be some under or overestimates on severity.

Four aspects of health service costs are monetised:

• physical injury health service cost
• emotional harm health service cost
• ambulance costs
• medical cost of a fatality

4.4.2 Physical health costs

Health costs are incurred from physical harms. To monetise these costs, assumptions are made regarding the treatment likely to be required for certain injuries based on their severity. The method uses IRS data, which records injuries by severity category and by type. The type of injury from the IRS definition (25 categories) is matched to the best possible proxy from NHS unit costs of injuries (NHS 2021). When assessing the right NHS unit cost, a weighted average of a variety of similar injuries is taken. Injuries classed as hospital slight take a weighted average of short stay, day cases and day admission costs for the injury type, whereas hospital severe injuries take the weighted average of day cases and elective/non-elective inpatients. The costs are weighted based on frequency as per the NHS data. As the IRS definitions are quite wide (such as “fracture”) a broad array of unit costs is taken and then averaged to get unit cost estimates.

Secondary and long-term injuries are calculated in a consistent way to section 4.1, with secondary injuries costed at one severity below the stated injury, with the assigned cost of treatment for the potential injury. For example, a spinal injury that appears to be serious would be accompanied by bruising, though this would be costed for treatment as bruising slight. Long-term costs for severe hospital admissions are considered, and it also assumed that 50% of these injuries have long-term (12-year) impacts, which decreases over time, as with section 4.1. The reduction over time accounts for an improvement in condition and consequently fewer health resources required to treat the injury, whether that is time or equipment. The long-term cost also accounts, albeit imperfectly, for there being any medical treatment outside of a hospital setting after an individual’s hospital admission. This cost is highly uncertain, but is expected to fall in the costs calculated below. The cost was also discounted appropriately over 12 years at 3.5% (HMT, 2022).

To estimate the health costs associated with the other harms, the unit cost of the treatment by severity is multiplied by the proportion of victims with that harm type at the given severity. Annex table A15 presents the unit cost of physical injuries by injury type, and table M21 presents total cost by fire type. The total cost is estimated to be £21.4 million.

Table M21: Total and unit health service costs from physical harms by fire type

Fire type	Total cost	Total injury incidents	Unit cost (Total incidents)	Unit cost (Primary incidents)	Unit cost (Injury incidents)
Other Buildings	£2.7m	313	£170	£190	£8,670
Dwellings	£14.1m	2,242	£450	£490	£6,290
Other Outdoors	£2.5m	188	£30	£470	£13,540
Road Vehicles	£2.1m	268	£100	£100	£7,680
All fire types	£21.4m	3,011	£140	£310	£7,110
Deliberate	£4.2m	488	£60	£220	£8,680
Accidental	£17.2m	2,523	£200	£350	£6,810

Table M21 notes:

1. Total costs rounded to the nearest £0.1 million, and unit costs rounded to the nearest £10.

4.4.3 Emotional health costs

Experiencing a fire related incident can lead to emotional trauma, which in some cases will require medical treatment. The cost of this medical treatment is proxied with an hourly cost of counselling (£49 from Curtis and Burns (2020)) and an estimate for the number of hours of counselling associated with each emotional harm. The same 3 emotional harms as 4.1.6 are used (fear, depression, and anxiety/panic attacks), with counselling hours of 2, 20 and 25 hours applied (HO, 2018) and therefore a unit cost of £98, £980 and £1,225.

As not all injuries will lead to an emotional harm, and not all emotional harms will require health service intervention, these unit costs are multiplied by the likelihood of an injury incident leading to emotional harms (matching 4.1.6, ONS, 2020C), and the likelihood of emotional harms requiring health service support which was gained from an NHS publication (39%, NHS, 2016). The latter figure (39%) is held constant regardless of whether the initial injury required hospital admissions or not, because once an individual has suffered emotional harm, the source is deemed to not impact severity. Long-term impacts are not considered in these emotional costs. The number of incidents that do not require hospital admissions (3,926), and the number that do require hospital treatment (3,011) are then multiplied by the relevant likelihood figures and unit costs to give total cost estimates, which are shown in annex tables A16 and A17. The total cost estimate is £0.9 million.

4.4.4 Fatalities and ambulance costs

Ambulances often respond to fire incidents, which also leads to a cost beyond the treatment cost already considered. It is assumed that an ambulance is required for all injuries that lead to a hospital admission and that the cost of an ambulance is £263 as per NHS (Curtis and Burns, 2020) unit costs for an ambulance seeing, treating, and conveying an individual. It is possible this slightly overestimates the total cost of ambulance response, as not all these injuries will necessarily require an ambulance. However, it may also slightly underestimate the total cost because ambulances may be mobilised to incidents and then not be required to convey an individual. The latter cost would not be included in this analysis. Multiplying the number of hospital admissions (3,011) by the unit costs gives £0.8 million.

There are also medical and ambulance costs associated with fire-related fatalities, which are additional to the cost of treating casualties already included. The unit cost of this is taken from the Tag data book (DfT, 2023, A4.1.1) in 2019 prices, to give a unit cost of £1,219. Multiplying this by the number of fatalities (245) gives a total cost of £0.3million.

4.4.5 Summary

Health costs from fire are presented, by type of health cost, in table M22 below. Table M23 gives a breakdown of these costs by fire type. The total health services costs are estimated to be £23.5 million.

Table M22: Total health costs by type

Type	Total cost	Unit cost (Total incidents)	Unit cost (primary incidents)
Physical health costs	£21.4m	£140	£310
Emotional health costs	£0.9m	£10	£10
Ambulance costs - injuries and fatalities	£1.1m	£10	£20
Total	£23.5m	£150	£340

Table M22 notes:

1. Total costs rounded to the nearest £0.1 million, and unit costs rounded to the nearest £10. Totals may not sum due to rounding.

Table M23: Total health costs by fire type

Fire Type	Total cost	Total fire incidents	Total primary fire incidents	Unit cost (Total incidents)	Unit cost (primary incidents)
Other Buildings	£2.9m	16,042	14,328	£180	£200
Dwellings	£15.6m	31,651	28,491	£490	£550
Other Outdoors	£2.7m	85,202	5,386	£30	£490
Road Vehicles	£2.2m	21,214	20,560	£100	£110
All fire types	£23.5m	154,109	68,765	£150	£340
Deliberate	£4.8m	69,850	19,159	£70	£250
Accidental	£19.8m	84,259	49,606	£230	£400

Table M23 notes:

1. Total costs rounded to the nearest £0.1 million, and unit costs rounded to the nearest £10.

4.5 Environmental costs

Fires attended by FRSs can have serious impacts on the natural environment, with wildfires and the emissions from fires posing a significant risk to England’s natural resources and human health. This report monetises this cost through carbon and particulate matter emissions and, where possible, the damage from outdoor fires. Outdoor fires, especially wildfires, are expected by many academics to increase in frequency and impact in the future, mainly because of climate change (Forestry Commission, 2014). Accordingly, these costs are expected to become more important and larger over time. As with all the other costs in this report, only the impact of fires attended by FRSs are included so any impacts or emissions from small “personal” outdoor fires not attended by FRSs (such as BBQs or campfires) are not in-scope of this analysis.

4.5.1 Carbon cost of emissions

One of the major impacts of fires on the environment is from their carbon emissions. The National Atmospheric Emissions Inventory (NAEI) within BEIS (2021b) publish annual estimates for the total carbon emissions from wildfires and other accidental fires (split into business and residential) in the UK (NAEI, 2021). Using this data, it is estimated that the total volume of emissions from fires is approximately 0.3765 million tonnes carbon dioxide equivalent (MtCO₂e) per year. The vast majority (0.37 MtCO₂e, 98%) is from wildfires. The remainder of the NAEI published emissions comes from “accidental business fires” (0.0027 MtCO₂e), which includes all emissions from other buildings fires, and “accidental residential fires”, (0.0038 MtCO₂e), which includes emissions from road vehicle and dwelling fires. These figures cover the whole of the UK, so are multiplied by 78% (the proportion of all UK fires attended that are in England) to get emission totals for England only^{[footnote 28]}.

The cost of these emissions can be calculated using the BEIS unit cost of carbon for year ending March 2020, which is £228 per tCO₂e (BEIS, 2021a)^{[footnote 29]}. This unit cost represents the monetary value that society places on one tonne of carbon dioxide equivalent and is used across government appraisal to allow for an objective, consistent and evidence-based approach to carbon valuation. It is important to note that these unit costs are calculated using a “target-consistent” or “abatement cost” approach, and so are related to the cost per tonne of reaching a net zero target, as opposed to the direct harm cost of carbon emissions. The unit costs also increase annually up to 2050, so should be adjusted in line with the BEIS estimates if used for appraisal in different years to year ending March 2020. The unit cost is multiplied by total emissions to estimate total costs. Costs are presented in annex table A18, and total approximately £67 million.

4.5.2 Impact on air quality

Fires can be a source of serious air pollution, which can impact air quality and individuals’ health. Academic literature has found that globally, on average, air pollution from wildfires is estimated to cause approximately 339,000 excess deaths a year (Marsh and McLellan, 2019). Studies of vegetation fires across Europe (including England), and the Saddleworth Moor fire have corroborated this, finding that emissions from fires can lead to excess deaths and negative health impacts on individuals (Kollanus et al., 2016; Graham et al., 2020).

To calculate the impact on individuals from reductions in air quality, NAEI estimates for total particulate matter (PM2.5) emissions by type of fire are used (NAEI, 2019). PM2.5 is an air pollutant which can have large negative impacts on people’s health. These emissions are monetised using the DEFRA damage cost estimates for PM2.5 (DEFRA, 2021a). This is estimated to be £68,000 per tonne in year ending March 2020 values and prices^{[footnote 30]}. This unit cost is a damage cost and mainly considers the impacts of PM2.5 on human health (health costs and wellbeing) but also considers environmental and economic costs (productivity). These impacts are distinct from the harms as a result of injuries discussed in sections 4.1, 4.3 and 4.4 as these harms are likely to be more long-term, and do not necessarily directly impact those at the scene of a fire. So, it is not expected that the IRS data on injuries will include these individuals. The unit costs are multiplied by total emission estimates to calculate total costs. These are shown in annex table A19 and total approximately £52 million.

These figures are tested using two academic reports into the number of excess deaths from outdoor fires. Kollanus et al., (2016) found that in 2005, approximately 52 premature deaths could be attributed to vegetation fire originated PM2.5 particles in the UK, with the equivalent number in 2008 being 42. These figures are converted to an equivalent figure in England year ending March 2020, accounting for the change in the number of outdoor fires over the period, and the number of fires in England compared to the UK (HO, 2021)^{[footnote 31]}. This gives an estimate of approximately 18 to 20 excess deaths caused by PM2.5 from outdoor fires in year ending March 2020, although this is potentially a slight underestimation (Kollanus et al., 2016)^{[footnote 32]}. Graham et al., (2020), who simulated the impact of the Saddleworth Moor fire, estimated that there were 8.6 expected excess deaths attributable to the fire and that it was 800 hectares in size. Uprating this fire to account for total outdoor fire damage in year ending March 2020 (3,708 hectares ^{[footnote 33]}, a multiplier of 4.6), and then only accounting for damage caused from the largest outdoor fires in the IRS (61% of damage is caused by outdoor fires that had individual damage of over 5000m²) to exclude small outdoor fires with limited emissions, gives a total estimate of excess deaths in year ending March 2020 of 24. Averaging out the three figures given by these two methods (18, 20 and 24) gives a central estimate of 21 excess deaths. Monetising these costs using the unit cost of a fatality from a physical harm, lost output, and health cost perspective (as outlined in sections 4.1.7, 4.3.3 and 4.4.4), gives a total cost of approximately £28.6 million, which aligns closely to the total outdoor cost estimate (£24.6 million) in annex table A20. This acts as strong corroborating evidence for the damage costs presented in table A20. However, this methodology may underestimate the full extent of the cost of poor air quality because it does not fully consider those who may suffer from respiratory conditions that have costs associated with them (health and QALY) but do not lead to increased mortality risks.

4.5.3 Damage from outdoor fires

Outdoor fires can lead to damage not considered within the property damage calculations in section 4.2. This damage can require additional response (from the National Trust and similar agencies), a need to restore and clean-up damaged land, and could potentially lead to wider indirect impacts (such as the loss of tourism and closure of roads) on the area impacted. These costs will vary largely by incident. This large amount of variation makes an accurate average unit cost difficult to calculate. Several case studies of outdoor fires were assessed to consider these costs, with the average across studies used where possible to account for variation.

To effectively uprate individual incidents to all outdoor fires, the total amount of damage caused by outdoor fires needed to be calculated. The Forestry Commission compared IRS data to woodland maps and Ordnance Survey data to classify fires into either wildfires or not. They analysed approximately 260,000 wildfire incidents between year ending March 2010 and year ending March 2017 inclusive (approximately 32,000 per year) and found that this led to approximately 37,000 hectares of land being burnt over the 8 years (Forestry Commission, 2019). This equates to approximately 0.14 hectares of damage per fire. Using IRS data on the approximate number of wildfires in year ending March 2020 (data sent to Forestry Commission) it was calculated that in year ending March 2020, there was approximately 3,700 hectares of damage from outdoor fires.

Response costs

Outdoor fires will require a level of response not incurred by other fire types. This will vary by fire, and there is limited evidence on the response costs that were incurred in year ending March 2020. The best available data comes from a case study of the Marsden Moor fire in 2021, which is used to estimate the total cost across all fires. This fire required staff from numerous organisations including the National Trust, Yorkshire Water, and the local council ranger service to deal with the response to the fire. This additional time on individuals has an economic cost. The National Trust also arranged for a helicopter to assist with the firefighting effort. The National Trust provided us with indicative figures for the cost of this additional response, and the helicopter. These were adjusted to make the employment cost assumptions consistent with the rest of this report, and then were uprated using amount of damage to provide a total cost of responding to all fires in year ending March 2020, not just this single event. As the Marsden Moor fire had a total burn area of 278 hectares, costs were multiplied by 13 to consider all fires. The helicopter cost was then multiplied by 61%, so it only accounted for damage from the largest fires, as we acknowledge that most outdoor fires are not of the size to require a helicopter response, although they may need some individuals to respond to clean up^{[footnote 34]}. The total cost of additional response to outdoor fires was therefore approximately £340,000 in year ending March 2020. This cost covers all outdoor fires within the wildfire Forestry Commission dataset (described above, figure which contributes to 3,700 hectares value). These fires include both primary fires (described as “other outdoors” in published Home Office statistics) and secondary fires. To ensure the costs in this section align with the rest of this report, this total cost is split into both primary and secondary fires according to total cumulative damage as per the IRS. Across the analysis, the split used aligns with the number of in-scope fires. So, for the additional response costs, a 41:59 per cent split between primary and secondary fires is applied, and for the helicopter costs, a 51:49 per cent split is used as only the largest fires are in-scope. The same principle is applied to split out accidental and deliberate fires. This leads to total primary costs of £160,000, and total secondary outdoor costs of £180,000.

Restoration and clean-up costs

Outdoor fires can also have considerable impacts on the land they occur on, impacting the natural environment. Some vegetation may be able to recover quickly from the fire, however other vegetation and habitats will require restoration work to recover. This can be very expensive after large incidents. One example of this is in moorlands, where fires can burn into the peat and lead to areas of bare peat, which require restoration. Other aspects of clean-up from outdoor fires include replacing damaged infrastructure and contents that could be lost in fires, such as fences, moorland dams, monitoring equipment in moors and forests, and other general infrastructure like gates, stiles and signs. These costs will also vary by incident, and so past case studies are used to estimate the potential size of these costs. Case studies of the Marsden Moor (2021), Bleaklow (2003), Flyingdales Moor (2003), Winter Hill and Stalybridge Moor (both 2018) fires are assessed, with restoration and clean-up costs taken from each of these incidents (Moors for the Future, n.d., and DEFRA, 2021b). These costs are converted to year ending March 2020 prices, uprated using multipliers of total damage of all incidents compared to the amount of damage from these incidents, and then adjusted to only include the largest fires, as not all outdoor fires necessarily require restoration. The average total estimate for restoration costs is then taken, leading to a total cost estimate of £3.0 million across all outdoor fires. This cost falls within a low and high range of £1.5 million to £3.7 million. The cost is split out between primary “other outdoor” fires and secondary fires using total cumulative damage within the in-scope outdoor fires. This leads to a total cost estimate of £1.5 million for primary “other outdoor” fires, and £1.4 million for secondary fires^{[footnote 35]}.

Total costs and wider indirect impacts

The total cost of response, restoration and clean-up across all outdoor fires is £3.3 million. However, it is acknowledged that wildfires, especially moorland fires, can have considerable wider impacts on the local area and agricultural sectors, especially when they occur in tourist areas or sites of special scientific interest. It has been estimated that the fires in Tameside in 2018, and Bleaklow in 2003, could have cost the local economy approximately £205,000 and £500,000 respectively (The Environment Agency, as cited by Moors for the Future, n.d.). As well as this economic impact and business interruption, large outdoor fires may also indirectly cause the closure of major roads, and transport hubs, which can have considerable economic impact (McMorrow et al., 2006). The scale of these wider indirect impacts is uncertain, lacking in significant research, and will vary by incident. To try and capture these within this report the cost of response, restoration and clean-up is uprated by 25%, which is broadly in line with the estimations for the 2003 Bleaklow fire (Moors for the Future, n.d.)^{[footnote 36]}. This assumption is highly uncertain.

The total cost of the wider indirect economic and social impacts is therefore £0.8 million, and the total cost estimate for damage from outdoor fires is £4.1 million. These costs are also split out between primary “other outdoor”, and secondary outdoor fires. The cost of wider impacts from primary fires is £0.4 million, giving a total cost of £2.1 million, and the cost of wider impacts from secondary fires is £0.4 million, giving a total cost of £2.0 million.

4.5.4 Total costs

The total environmental costs are therefore estimated to be £123 million across these cost areas, equating to a unit cost of £800 across all 154,109 fires attended. Table M24 below gives the total and unit environmental cost by fire type. Further detail is presented in annex table A20.

Table M24: Total and unit environmental costs by fire type

Fire Type	Total cost	Unit cost (Total incidents)	Unit cost (primary incidents)
Other Buildings	£7.0m	£430	£490
Dwellings	£8.0m	£250	£280
Other Outdoors	£94.5m	£1,110	-
Of which are primary	£39.4m	£7,310	£7,310
Of which are secondary	£55.1m	£690	-
Road Vehicles	£13.3m	£630	£650
Total	£122.8m	£800	£1,790
Deliberate	£48.7m	£700	£2,540
Accidental	£74.1m	£880	£1,490

Table M24 notes:

1. Total costs rounded to the nearest £0.1 million, unit costs rounded to the nearest £10.

4.6 Insurance Administration

There are also economic costs incurred by insurance companies from the administration and management of claims made by individuals, once a fire has occurred. This is a cost to society because these resources could have otherwise been employed in alternative productive activities, and it is included in the consequence section because this cost would be reduced if the number of fires were to fall.

To estimate the cost of insurance administration related to fire claims, data from the Association of British Insurers (ABI) on the total value of claims management expenses and insurance claims incurred is used to estimate the costs directly associated with processing and resolving insurance claims. The ABI data is broken out into domestic and commercial costs, and both are included. The data is then adjusted to account for proportion of insurance administration costs which are expected to be fire related, and the amount of market coverage that the ABI have, so a full England cost can be calculated.

Total insurance administration expenditure in year ending March 2020 was estimated to be approximately £114 million. This cost can be broken down into dwelling fire related expenditure (£42 million) and other building related expenditure (£72 million).

4.7 Wider consequences

4.7.1 Crime specific costs

Some deliberate fires will be criminal offences and occur as a result of criminal behaviour, so therefore incur costs from police and criminal justice system (CJS) action, as well as incurring additional costs from victim services. There are also costs in anticipation of arson (see section 3.3). The approach to calculating these costs in response to arson was taken from the economic and social cost of crime (ESCC, 2018) report (HO, 2018), with the unit CJS and victim service costs uprated with the GDP deflator (HMT, 2023). The police costs were calculated by adjusting the ESCC 2018 police unit costs (HO, 2018), to account for the year ending March 2016, police funding, and then multiplied by the uplift in the police budget for England between year ending March 2016, and year ending March 2020. These unit costs are then multiplied by the number of arson offences that occurred in England in year ending March 2020. This gave total crime specific costs of deliberate fires to be £78 million. This is split into £16 million for police services, £62 million on criminal justice services, and £0.2 million on victim services (see annex table A21).

As there is no data available on arson offences by location in either the crime survey or police recorded crime, it was assumed that crime specific costs occurred proportionally to primary deliberate fire incidents. Therefore, it was assumed that the unit cost of all primary deliberate incidents was equal, across fire types. The unit cost of a primary deliberate fire will be lower than an arson offence, because not all primary deliberate fires will be arson offences (some will have a unit cost of 0). Table M25 presents estimates for unit and total cost by fire type. Only deliberate incidents are included because it was assumed that all criminal arson offences would be classed by FRSs as deliberate fires, not accidental fires.

Table M25: Crime specific costs by fire type

Fire Type	Total cost	Total deliberate incidents (All)	Total deliberate primary fire incidents	Unit cost (Total incidents)	Unit cost (Primary incidents)
Other Buildings	£16m	5,091	3,880	£3,100	£4,000
Dwellings	£12m	3,194	2,969	£3,800	£4,000
Other Outdoors	£10m	51,257	2,471	£200	£4,000
Road Vehicles	£40m	10,308	9,839	£3,900	£4,000
Total (all deliberate)	£78m	69,850	19,159	£1,100	£4,000

Table M25 notes:

1. Total costs are rounded to the nearest £1 million, and unit costs are rounded to the nearest £100.
2. Assumed that no secondary fires lead to crime related costs.

4.7.2 Firefighter injuries

Firefighter injuries are costed the same way as injuries to other victims of fire. The cost of physical and emotional harm is included, as well as additional costs on health services. The costs are calculated using IRS data, in which firefighter injury data is collected in the same way as data on other victims of fires. There were 124 injuries to firefighters in year ending March 2020, which are assessed in this analysis and 4 cost areas are included. These are physical harms (QALY impact) totalling £0.2 million, emotional harms (QALY) totalling £0.1 million, physical health services costs totalling £0.1 million and emotional health services costs totalling approximately £20,000. The total cost of firefighter injuries is therefore £0.4 million, with full data given in annex table A22.

Wider potential harms to firefighters resulting from PTSD or other long-term impacts has not been costed due to the unavailability of a metric by which to measure it. This would be an area for further research.

4.8 Totals

4.8.1 Total overall costs

The total consequence costs as a result of fire are estimated to be £3.15 billion, and are presented in table M4 earlier in this report. Total and unit costs are presented, with the unit cost split into primary incidents and all incidents. Property damage is the largest cost area at approximately £2.0 billion (64%) of total costs. This is followed by physical and emotional harms (£362 million, 11% of total), road vehicle damage (£272 million, 9%), and lost output (£170 million, 5%).

A full breakdown of consequence costs by fire type are presented in annex table A23. There are significant cost drivers of some fire types, with property damage making up 85% of the cost of “other building” fires, and road vehicle damage making up 76% of the cost of road vehicle fires. The unit costs by fire type are also presented in annex table A24.

4.8.2 Accidental versus Deliberate Fire totals

The consequence costs are also split out by accidental and deliberate fires. The unit cost is presented when only considering primary fires attended, and when considering all fires attended.

The unit costs of accidental and deliberate primary fires are similar, at £44,000 and £47,700 respectively. The unit cost of all deliberate fires is significantly lower (£13,400 compared to £26,300 for accidental fires), driven by the large number of secondary outdoor fires with limited consequences. For primary fires, the unit consequence costs of deliberate dwellings, other buildings, other outdoor and road vehicle fires are larger than the same unit costs for accidental fires. For dwellings, there is a notable difference (accidental fires are £28,400, whereas deliberate fires are £59,600). The unit cost of deliberate fires is seen as a good proxy for an arson offence. However, the total cost should not be used as an estimate for the total cost of arson because of differences in the number of recorded arson and deliberate fire offences. A full breakdown of the consequence costs of accidental and deliberate fires by fire type is presented in annex table A25.

5. Costs in response to fire

This chapter will discuss the methodological approach to calculating the costs in response to fire and then present total cost estimates. The cost is solely that of FRSs responding to fire incidents. The cost of other agencies (such as police, CJS and health service) responding to a fire incident is included in the consequence section of this report, as it is deemed that these costs are more driven by the consequences of fires, as opposed to fires happening. The cost of FRSs responding to fire false alarms and all non-fire incidents is also calculated in this section. However, this is considered to be out of scope of the overall estimate for the cost of fire. There is also some consideration of other non-monetised wider costs in response to fire.

5.1 FRS costs

This section calculates the cost to FRSs from directly responding to fires. This cost will include the cost of firefighters responding to incidents, the cost of control staff facilitating response, and non-labour costs directly incurred from response. This section does not equate to all FRS expenditure (which includes variable and fixed cost), but just the direct cost of the service responding to fires (variable cost). These are costs that would be reduced or incurred if the number of fires were to fall or rise.

5.1.1 Direct labour costs (firefighters)

To monetise the direct cost of FRSs responding to incidents with firefighters, incident recording system (IRS) data was used. The IRS collects data on all incidents attended by FRSs, and all the appliances mobilised to these incidents. The appliance level data contains date-time stamps for several key components of incident response, including:

• time appliance is mobilised for an incident
• time appliance arrived at an incident
• time appliance is available following an incident (it has finished responding)

These date-time stamps can be used to calculate the total time that each appliance in each FRS spent attending fire incidents in year ending March 2020. For each incident attended, each appliance’s drive time is added to their total attending time to account for the time it takes an appliance to return to the station post-incident, and/or any other required debrief time^{[footnote 37]}. This is considered as part of the total economic cost of response, as this time could not feasibly be spent by individuals on the appliance undertaking other productive activity, and if the appliance had not been mobilised, this is additional time the individuals on the appliance could have done something else productive. Figure 1 below explains how the mobilised time for each appliance is calculated, based on IRS data.

Figure 1: Calculation of total mobilised time for each FRS appliance

Further detail on response times can be found in the Home Office published response time data. The published response time data only assesses the first appliance to respond to each incident. However, to capture the whole cost of fire, the data for this report includes all appliances that responded to any fire incident in year ending March 2020. The time between “time of call”, and “time appliance mobilised”, referred to as “call handling time”, is not considered to be part of total appliance mobilised time because during this time firefighters could be undertaking alternative activity, and are not yet beginning to respond to the incident. This only occurs from the point of appliance mobilisation.

Using IRS data and this method to calculate mobilised time per incident, total appliance mobilised time for every service, per appliance type, could be calculated. However, in order to accurately cost the time FRSs spend responding, data on the number of crew and their role (and therefore salary) was needed. This data is not included in the IRS, so to collect this data, a crewing survey was launched in 2021 to every FRS^{[footnote 38]} which asked them for data on:

• average crewing numbers by role and appliance type
• information on the non-labour costs associated with running appliances
• average employment costs for control and other operational staff (wage plus non-wage costs)

This crewing and salary data could be combined with mobilised time data, to estimate the total direct labour cost associated with response. Where services did not respond, average crewing numbers and salaries were applied. A worked example of the method used for one incident is presented below for full clarity:

A pump ladder appliance was mobilised for 45 minutes to a fire (sourced from IRS).

There were 4 Firefighters on the appliance (sourced from crewing survey).

Their hourly wage was £18 (indicative for example, actual data used came from crewing survey).

Total cost of this appliance responding therefore: 4 x £18 x 0.75 = £54.

This method is repeated for every appliance mobilised in the IRS, for all fire, fire false alarms and non-fire incidents attended. It was assumed crewing would be equivalent across these incident types, as crewing is typically linked to the appliance mobilised. The total costs were calculated on an individual FRS level, to ensure the highest level of accuracy possible, with costs summed and aggregated costs presented in table M26. Although fire incidents are the lowest in number, they are the highest average cost and represent 47% of total direct labour costs. This is because they are typically responded to for longer, and by more appliances, than other incident types.

Table M26: Direct labour costs (firefighting) from FRS response by incident type

Direct labour costs	Total cost	Total incidents	Unit cost (Incidents attended)	Proportion of total cost
Fire	£30.4m	152,289	£200	47%
Fire false alarms	£14.4m	231,052	£60	22%
Non-fire	£19.7m	169,687	£120	31%
Total	£64.4m	553,028	£120	100%

Table M26 notes:

1. Only incidents with full appliance data included (so incident numbers marginally different to those elsewhere in report). Total costs rounded to the nearest £0.1 million and unit costs rounded to the nearest £10.

5.1.2 Direct non-labour costs

The direct labour costs described so far do not represent the full cost of responding to fires, as they do not account for any of the direct non-labour costs associated with FRS response. These include the cost of fuel, maintaining and servicing appliances due to wear and tear, and other costs such as tyres, insurance, road tax, and leasing vehicles (in some services). These costs were collected from services as part of the crewing survey. Where services did not respond to the survey, average non-labour costs, as a proportion of total direct labour costs were applied. This average, when weighting services according to their total direct labour costs, was estimated to be 72%^{[footnote 39]}. These non-labour costs covered responding to all incidents (of any type), as many services could not break the costs down by fire and non-fire incidents and totalled approximately £46.5 million. These non-labour costs were split out by incident type according to total direct labour costs as it was assumed that non-labour cost expenditure is broadly in line with labour cost. The more an appliance is used, and the more the labour cost of running it is, the more non-labour cost is likely to be incurred. Table M27 shows the total non-labour cost, broken down by incident type.

Table M27: Direct non-labour costs from FRS response by incident type

Direct non-labour costs	Total cost	Total incidents	Unit cost (Incidents attended)
Fire	£21.7m	152,289	£140
Fire false alarms	£10.3m	231,052	£40
Non-fire	£14.5m	169,687	£90
Total	£46.5m	553,028	£80

Table M27 notes:

1. Total costs rounded to the nearest £0.1 million and unit costs rounded to the nearest £10.

5.1.3 Control costs

The final aspect to consider for FRSs response to fires is the cost of control staff. These are uniformed staff working in FRS control centres to answer emergency calls and deal with mobilising, communications, and other related activities^{[footnote 40]}. To monetise the cost of control staff, crewing survey data on the average salary of control staff for each FRS at each role was multiplied by the number of control staff at each role, sourced from Home Office published data (HO, 2023, FIRE1102) . Where services share control functions (and therefore have 0 control staff marked in the Home Office data), the total cost of the control function was split out according to total direct non-labour and labour costs, or more specific data if this was provided in the crewing survey. This approach was taken for a number of FRSs and control centres, such as North West Fire Control, which is split between Cheshire, Cumbria, Lancashire and Greater Manchester, and the Berkshire control room which is shared with Oxfordshire and Buckinghamshire. Where FRSs or control centres did not respond to the crewing survey, average employment costs were applied^{[footnote 41]}. The total control cost was calculated to be £46.7 million. The control cost in each FRS was split out between fire, fire false alarm, and non-fire spend, according to total direct labour costs as it was assumed that control time is broadly in line with labour cost. All control cost is included in the cost of response as unlike firefighter time, or non-labour expenditure, all control time is assumed to be spent on direct response related activity. Further work could be completed to refine this assumption. Total control costs, broken down by incident type, are in table M28.

Table M28: Control costs from FRS response by incident type

Control costs	Total cost	Total incidents	Unit cost (Incidents attended)
Fire	£21.8m	152,289	£140
Fire false alarms	£10.2m	231,052	£40
Non-fire	£14.7m	169,687	£90
Total	£46.7m	553,028	£80

Table M28 notes:

1. Total costs rounded to the nearest £0.1 million and unit costs rounded to the nearest £10.

5.1.4 Total costs of FRS response and discussion

Direct labour, non-labour costs and control costs, were summed to calculate the total cost of FRS response. Only the cost of fire related response is included in the total cost of fire. However, the cost of fire false alarms and non-fire incidents is presented in this report and used to calculate the cost of FRS fire related anticipation spend (see 3.7)^{[footnote 42]}. The total cost of response by incident type is presented in table M29.

Table M29: Total response costs by incident type

Total response costs	Total cost	Total incidents	Unit cost (Incidents attended)
Fire	£73.9m	152,289	£490
Fire false alarms	£34.9m	231,052	£150
Non-fire	£48.8m	169,687	£290
Total	£157.7m	553,028	£290

Table M29 notes:

1. Total costs rounded to the nearest £0.1 million and unit costs rounded to the nearest £10.

As a sense check, the total cost of response was compared to total FRS revenue expenditure in year ending March 2020, £2.2 billion (MHCLG, 2020), to estimate that the cost of directly responding to incidents is 7.1% of FRS revenue expenditure. The total cost of fire response was also split out by fire incident type. This was done by calculating the direct labour cost for each incident type in the IRS by FRS and then uprating these to account for non-labour and control costs, with the assumption that these costs are spread across incident types proportionally to labour costs. Total estimates, by incident type, are in table M30.

Table M30: Total response costs by fire incident type

Total response costs	Total cost	Total incidents	Unit cost
Chimney	£1.4m	3,080	£470
Dwellings	£21.3m	28,092	£760
Other Buildings	£23.2m	14,129	£1,640
Other Outdoors	£7.6m	5,295	£1,440
Road Vehicles	£6.5m	20,315	£320
Secondary Fires	£13.9m	81,378	£170
Total	£73.9m	152,289	£490
Deliberate	£21.3m	69,240	£310
Accidental	£52.7m	83,049	£630

Table M30 notes:

1. Total costs rounded to the nearest £0.1 million and unit costs rounded to the nearest £10.

There is a notable difference in the unit cost of FRSs responding to incidents of different types, with “other building” fires having the highest unit cost of response, and secondary fires having the lowest. There is also a large difference between the average cost of responding to deliberate and accidental fires. The differences are caused by different lengths of time attending incidents (secondary and chimney fires are typically very low damage, so less time is required to attend these incidents), and different levels of appliance attendance. Each of the unit costs presented in table M30 vary by service, which will be caused by differing incident sizes, appliance mobilisation levels, average employment costs, and non-labour expenditure.

The unit cost of fire false alarms is lower than the response to any type of fire, which is unsurprising due to the different lengths of attendance and number of appliances mobilised between a fire and false alarm incident. It is notable that although fire false alarms represent the largest number of incidents attended by FRSs, they have the lowest total costs (table M29). However, this cost still represents a substantial burden on FRSs.

5.2 FRS cost in anticipation of fires

As well as being used to calculate total response costs, the split of direct labour costs is used to calculate the total expenditure by FRSs in anticipation of fires (section 3.7). FRSs spend a large amount of their time and resource training to be able to respond to fires, ensuring resilience to fires, and on prevention and protection activity that aims to minimise the occurrence and impact of fires. As all this expenditure is directly related to fire, it should be included in the total cost of fire. However, it is better classed as anticipation related spend instead of response related spend, due to the spend being incurred “before” a fire event instead of in response.

It would not be correct to simply remove the total response cost of all incident types from total FRS expenditure in year ending March 2020 (£2.2 billion) to estimate this cost, as this does not account for the anticipation related activity that FRSs complete for non-fire activities. To account for this correctly, total FRS spend not related to response (approximately £2.0 billion) is multiplied by the proportion of direct labour costs incurred from non-fire response (31% on average across all services, see table M26). This is completed for each FRS, and when summed across all services, equals £618 million. The remaining spend, £1.4 billion, is classified as expenditure by FRSs on anticipation^{[footnote 43]}. Figure 2 below shows how total FRS expenditure is distributed, with table M31 showing total and unit costs. The vast majority of FRS spend is incurred on anticipation related activity. All this cost is captured in the anticipation cost section of this report (section 3.7).

Figure 2: Total FRS expenditure by type (year ending March 2020). Costs in £million

Table M31: Total FRS expenditure by type

Total response costs	Total cost	Total incidents	Unit cost
Fire response	£73.9m	152,289	£490
Fire False alarms	£34.9m	231,052	£150
Non-fire response	£48.8m	169,687	£290
Non-fire anticipation	£616.6m	169,687	£3,630
Fire anticipation	£1,384.2m	152,289	£9,090
Total fire	£1,458.1m	152,289	£9,570
Total non-fire	£665.4m	169,687	£3,920
Total FRS expenditure	£2,158.5m	-	-

Table M31 notes:

1. Total costs rounded to the nearest £0.1 million and unit costs rounded to the nearest £10.
2. Total fire includes fire response and fire anticipation. Fire false alarms are not included.

Further work could be completed to refine these costs. The cost of non-fire response could be disaggregated into different non-fire activities. However, this was viewed as beyond the scope of this report. Similarly, the cost of fire anticipation spend could be split into different FRS outputs such as prevention and protection activity. This additional detail could be built into future reports, but it was not possible in this report with the available data. It is important that when using these anticipation costs, it is clear that they include all non-response fire related activities undertaken by FRSs. The spending on these activities is not equivalent to just the spending on prevention, protection, or community safety activity, as the cost also includes the resilience of services, training, back office functions, and fixed costs that are necessary to have a functional FRS.

5.3 Wider non-monetised cost of response

There are some further areas of the response to fire incidents, which have not been monetised in this report, but would represent interesting areas for future research. The cost of other agencies responding to fires (such as Police) is included in the consequence section of this report.

Central call handling:

There are likely some economic costs associated with the central call handling of 999 calls following fires, which are not included in current control costs above. A direct 999 call will usually be answered by central telephone exchange operators, employed by BT, who then pass the call onto the relevant fire control centre. This time and expenditure is an economic cost of fire. However, there is currently limited data on the scale of this cost.

Industrial fire brigades:

Industrial, or private, fire brigades operate in England in some locations such as airports, Ministry of Defence sites, and other private locations such as factories. These are different to the English FRSs referenced throughout this report and are sometimes operated by private companies. As these services respond to fire incidents, this cost is considered a cost of fire response. Although the incidents responded to are not necessarily also responded to by FRSs, and so do not meet the definition of an incident attended by an FRS, without the existence of these brigades and services, it is expected that FRSs would have had to respond or respond earlier (or with more appliances).

This therefore still represents an economic cost of fire that should be included in this report, as in the absence of the alternative fire service, these costs would have been incurred by English FRSs in response. However, limited data exists on the scale and cost of these brigades, so they are not costed in this report. Further work could be completed to estimate their total cost.

6. Future developments

This economic and social cost of fire report is the best available current estimation and is a significant addition to the fire economics evidence base. It provides an up-to-date robust and reliable estimate for the impact of different fire types in England.

However, there are still some uncertainties in the report, and there are a number of further developments and enhancements that would improve the estimations. These are either considered to be beyond the scope of this report or were not possible due to a lack of available data. They are expected to form the basis for the next update of the report and are areas where future external research and analysis could focus. These areas are listed below and split by area of the report.

Anticipation

Improving defensive expenditure estimates

This report has presented estimates for defensive expenditure on buildings and consumer goods, which make up a substantial proportion of the total economic and social costs of fire (£6.6 billion of £12.0 billion, 55%). These costs have been calculated using the best available data. However, further work could be done to refine them and collect additional data to improve the estimates. The latter could be done by approaching and surveying organisations to calculate their average expenditure. Further research could also look to split these costs out further by fire type, and understand how they change over time, considering the impacts of recent fire safety regulations and to consider additional spend on cladding remediation. The cost of consumer goods could also be improved to better consider the differences between the studies used and England manufacturing, and to account for potential net imports.

Splitting out the cost of FRS anticipation spend

The current estimate for FRS anticipation spend is approximately £1.4 billion. Further research could be completed with FRSs to split this into more specific activity and different components, for example the cost of prevention, protection, training, and resilience activity, to better understand what this cost consists of. This could allow for activity based costing, and a better understanding of the cost of fire for FRSs. Similar work could be completed for the non-fire anticipation cost.

Consequence

Monetising wider consequences of fire

There are a number of consequence costs that could not be accurately monetised and therefore are not included in the total costs in this report. Future research could look to monetise these costs, and some of the non-monetised consequence cost areas that were identified through this project, are outlined below:

1. Evacuations: As a result of fires, individuals may be injured, require rescue by FRSs, or evacuate the property or building they are in. The former 2 impacts are costed in the physical and emotional harms section of this report. However, the latter, evacuations, is not. This is because of a lack of available data, and uncertainty on the impact of those who are evacuated from buildings due to fires. There are 2 aspects to this cost which could be considered; the time cost to individuals from evacuating buildings, and emotional harms.
2. Fear of fire: The emotional harm, including fear, of those who are injured or rescued from fires is monetised in the physical and emotional harm section of the report. However, many individuals who have not been involved in a fire (injured, rescued, or evacuated) are still fearful of fires in their homes. This can impact their quality of life, and lead to emotional harms.
3. Wider environmental harms: Outdoor fires, specifically wildfires, can have considerable negative impacts on wildlife, habitats, and biodiversity, which is not monetised in this report. They may also impact water quality (if residue from fires enters reservoirs or water sources), although this is not unique to wildfires, and the water used to fight fires can have large environmental impacts if not controlled and cleared. Equally the extinguishing and disposing of waste materials following a building fire can take extra time and effort to do safely, and the removal of unsafe structures can be costly. Outdoor fires may also have wider air quality impacts from other toxic gasses and smoke not monetised in this report.
4. Relocation: There may be costs from relocating individuals whose properties are damaged following a fire. This could involve expenditure on emergency accommodation when required, and travel costs to alternative accommodation (which can include opportunity time cost, and financial costs).
5. Community impacts: Fires can have extremely detrimental impacts on the local community they occur in. These business and employment impacts can be larger than those monetised within this report. As this report looks at business impacts on a national level, it accounts for the transfer of business between organisations. A fire may shut down a business, however the impact on society may not be as large as the impact on an individual because another business gains from this loss. There will still be large non-monetised impacts on individuals and communities if important businesses or other community sites are damaged by fires, but these harms have not been monetised.
6. Heritage and cultural costs: Fires could have large costs if they impact specific buildings of local community, heritage, or cultural importance and value. These buildings include places of worship or schools. Some noticeable examples include the 1992 Windsor Castle fire, and the 2019 Notre-Dame fire. Fires in heritage sites can lead to unique impacts on the local area, beyond those included in this report: for example uniquely large rebuild costs and wider economic losses such as a reduction in tourism. These vary by incident, but as an example the Windsor castle fire (1992) has been estimated to have cost approximately £36.5 million, and took 5 years to rebuild (National World, 2022). The Notre Dame fire (2019) raised approximately €750 million to rebuild it (BBC News, 2019).
7. Wider caring aspects: There is also likely to be a cost to children or family carers who are required to care for people injured in fires. This has an opportunity cost as this time would either be spent in education (in the case of children), or in undertaking other productive activities (for other carers).
8. Transport impacts: There can be potentially large transport related costs from fires occurring. These include the cost of fires in transport hubs (such as airports and rail stations), which can lead to delays to individuals (both directly and indirectly) and additional, specific damage costs. There can also be wider transport impacts, outside of fires in these locations. Incidents near to roads and motorways may lead to them being shut for a period of time, which could cause travel disruptions and delays. This can have an economic cost in terms of time spent, with individuals being unable to work or having to drive further distances to avoid road closures, which has an opportunity cost.

Improving property damage estimates

The cost of property damage estimates is the largest in the consequence section of the report. However, they have significant uncertainty associated with them, as numerous assumptions have had to be made for them to be calculated. Further work could look to improve on the assumptions in this document, for example by using more specific BCIS or wider property damage estimates.

Improving long-term injury and harm estimates

The cost, duration, and impact of long-term harms is highly uncertain, due to limited academic evidence, and high variation across injury types and the circumstances of harm. Currently, high-level assumptions are made in this report, which are based on the burns injury literature. Future analysis could be completed to improve these assumptions, potentially through further analysis of health data, and even linking IRS and fire injury data to hospital and other health data. Long-term health costs could also, in future, better account for medical treatment outside of a health setting, which are currently imperfectly calculated and may be underestimated. The assessment of health and QALY costs also does not account for the varying distribution of impacts across different individuals and groups.

Response

Monetising additional response costs

The response chapter of this report references additional potential response related costs, including central call handling cost and the cost of industrial fire brigades. These have not been monetised in this report. However, further research could look to cost them.

Improved estimates for non-fire response activity

The scope of this report is to focus on fire incidents, and their impact on FRSs. However, as part of this calculation work has been completed to calculate the average FRS response costs to non-fire incidents. These figures are based on the time to respond to non-fire incidents in the IRS, and average crewing from the crewing survey. Further research in this area could look in more detail at FRS response to non-fire incidents, potentially splitting the costs out by response type (for example the response costs to road traffic collisions compared to flooding incidents). This would be a useful and interesting addition to the fire evidence base and could improve the estimates in this report. However, it is likely the overall cost of non-fire response (which is used in this report to help determine fire anticipation costs), would be relatively unchanged.

Refining the control cost estimate

Currently, this report includes the full cost of control staff in fire response costs. This is seen as the best assumption with available data, however further work could be completed to refine this assumption and potentially reduce it from the current assumption that all control time is spent directly on supporting response so that time spent training or on other activities is considered, in the same way as firefighter time is assessed. This would require the collection of additional data, and without this, it is considered an accurate assumption to assume that all of the time available to control staff can be directly attributable to response.

Overall

Further consideration of fires not attended by fire and rescue services

The scope of this report is to consider fires attended by FRSs, and this has a clear rationale, and ensures that the estimates can be compared to IRS data. Fires not attended by FRSs are likely to be low impact and have limited consequences. However, further work could be undertaken to understand what the cost and impact of these fires is. It is possible that some small fires could lead to injuries and damage, even if they are put out by individuals and do not lead to an FRS response. Future research could test this hypothesis and better understand the scale and potential impact of fires without an FRS response.

Expanding the unit and total costs into other specific estimates

Currently this report estimates the unit cost of fire across a plethora of cost areas and splits these costs by fire type and whether the fire is accidental or deliberate. This is a large amount of detail. However, further work could be completed to further split out the unit costs into cause and / or more specific location types such as dwelling type. This may be difficult to do in a robust way due to some small sample sizes once fires are considered at a granular level. However, this could be overcome in future research with additional data and by using multiple years of data.

Widened Home Office data collection

Most of the analysis in this document is based on IRS data, and so the analysis in this document could be improved with a wider collection of incident data. Examples of improvements to the collected data that could assist this analysis would be a better consideration of overall response time by FRSs (an inclusion of time returning to stations, and post-incident debrief time), the inclusion of FRS activity data beyond response (for example time spent on prevention and protection activity, or training), and improved data on the nature of property damage as a result of fire (for example what items were impacted, and where exactly in a building the fire occurred). The collection of this data will be considered as part of data review for implementation after the procurement of the new National Fire Data Collection System.

7. References

Allied Market Research (2020), ‘Fire Protection Systems Market by Product Type, Type, and End user: Global Opportunity Analysis and Industry Forecast, 2020-2027’, https://www.alliedmarketresearch.com/fire-protection-system-market.

AMA Research (2021), ‘Active Fire Protection Market Report - UK 2020-2024’, https://www.amaresearch.co.uk/report/active-fire-protection-market-report-uk-2020-2024/.

Ashe, B., McAneney, J., and Pitman, A. J. (2009), ‘Total cost of fire in Australia’, Journal of Risk Research, 12(2): 121-136.

Association of British Insurers, ABI (n.d.), ‘Fire, what is it?’, https://www.abi.org.uk/products-and-issues/choosing-the-right-insurance/home-insurance/fire/.

BBC News (2019), ‘Notre Dame fire: Has too much money been given to rebuild it?’.

Centre of Economics and Business Research, CEBR (2014), ‘The financial and economic impact of warehouse fires’.

Choinière, M., Melzack, R., Papillon, J. (1991), ‘Pain and paresthesia in patients with healed burns: an exploratory study’, Journal of Pain and Symptom Management, 6(7): pp. 437-444.

Curtis, L. A., Burns, A. (2020), ‘Unit Costs of Health and Social Care’, PSSRU, University of Kent.

Dauber, A., Osgood, P. F., Breslau, A. J., Vernon, H. L., Carr, D. B. (2002), ‘Chronic persistent pain after severe burns: a survey of 358 burn survivors’, Pain Medicine, 3(1): pp. 6-17.

Department for Business, Energy & Industrial Strategy, BEIS (2021a), ‘Valuation of greenhouse gas emissions: for policy appraisal and evaluation’.

Department for Business, Energy & Industrial Strategy, BEIS (2021b)^{[footnote 44]}.

Department for Communities and Local government, DCLG (2011a), ‘The economic cost of fire: estimates for 2006’, Fire research report 2/2011.

Department for Communities and Local government, DCLG (2011b), ‘The economic cost of fire: estimates for 2008. Fire research report 3/2011’.

Department for Communities and Local Government, DCLG (2015), ‘English Housing Survey: Fire and Fire Safety 2013-14’.

Department for Environment, Food & Rural Affairs, DEFRA (2021a), ‘Air quality appraisal: damage cost guidance’.

Department for Environment, Food & Rural Affairs, DEFRA (2021b), ‘England Peat Action Plan’.

Department for Transport, DfT (2023), ‘Tag data book’, 2019 price year and 2019 value year used.

Dolan, P., Peasgood, T. (2007), ‘Estimating the Economic and Social Costs of the Fear of Crime’, The British Journal of Criminology, 47: pp. 121-132.

Esselman, P. C., Askay, S. W., Carrougher, G. J., and Lezotte, D. C., Holavanahalli, R. K., Magyar-Russell, G., Fauerbach, J. A., Engrav, L. H. (2007), ‘Barriers to Return to Work After Burn Injuries’, Archives of Physical Medicine and Rehabilitation, 88(12 Suppl 2): pp. S50-S56.

Fire Protection Association (2020), ‘FPA launches Know your Building campaign as large fire losses average £650k’.

Forestry Commission (2014), ‘Building wildfire resilience into forest management planning’.

Forestry Commission (2019), ‘Wildfire statistics for England’.

GBD (2019), ‘Global Burden of Disease Study 2019 (GBD 2019) Disability Weights’.

GOV.UK (2015) ‘Incidents of fires in the home’.

Graham, A. M., Pope, R. J., Pringle, K. P., Arnold, S., Chipperfield, M. P., Conibear, L. A., Butt, E. W., Kiely, L., Knote, C., McQuaid, J. B. (2020), ‘Impact on air quality and health due to the Saddleworth Moor fire in northern England’, Environmental Research Letters, 15(7).

HM Treasury, HMT (2022), ‘The Green Book: Central government guidance on appraisal and evaluation’.

HM Treasury, HMT (2023), ‘GDP deflators at market prices, and money GDP’.

Home Office, HO (2018), ‘The economic and social costs of crime, Second edition’.

Home Office, HO (2019), ‘The economic and social costs of domestic abuse’.

Home Office, HO (2021), ‘Detailed analysis of fires attended by fire and rescue services, England, April 2020 to March 2021’.

Home Office, HO (2022a), ‘Impact Assessment, The Home Office: Emergency Evacuation Information Sharing (EEIS)’.

Home Office, HO (2022b), ‘Fire statistics definitions’.

Home Office, HO (2022c) ‘Fire and rescue incident statistics: Methodology and quality report’.

Home Office, HO (2023), ‘Fire statistics data tables’.

Homes and Communities Agency, Kirklees (2015), ‘Employment Density Guide, 3rd Edition’.

James, S. L., et al., (2020) ‘Epidemiology of injuries from fire, heat and hot substances: global, regional and national morbidity and mortality estimates from the Global Burden of Disease 2017 study’, Injury Prevention, 26: pp. i36-i45.

Kollanus, V., Prank, M., Gens, A., Soares, J., Vira, J., Kukkonen, J., Sofiev, M., Salonen, R. O., Lanki, T. (2016), ‘Mortality due to Vegetation Fire-Originated PM2.5 Exposure in Europe – Assessment for the Years 2005 and 2008’, Environmental Health Perspectives, 125(1): pp. 30-37.

London Fire Brigade, LFB (2022), ‘Fire safety training’.

Macro Trends (2022), ‘U.S. Manufacturing Output 1997-2023’.

Marsh and McLellan (2019), ‘The Burning Issue: Managing Wildfire Risk’.

McMorrow, J., Aylen, J., Albertson, K., Cavan, G., Lindley, S., Handley, J., Karooni, R. (2006), ‘Moorland Wild Fires in the Peak District National Park, Technical Report 3. (Climate Change and the Visitor Economy (CCVE))’, University of Manchester, Centre of Urban and Regional Ecology.

Roofing Today (2018), ‘Integrated systems stimulate growth in the UK active fire protection sector’.

Ministry for Housing, Communities and Local Government, MHCLG (2020), ‘Revenue Outturn central, protective and other services (RO6) year ending March 2020’.

Moi, A. L., Haugsmyr, E., Heisterkamp, H. (2016), ‘Long-Term Study of Health and Quality of Life After Burn Injury’, Annals of Burns and Fire Disasters’, 29(4): pp. 295-299.

Moors for the Future (n.d.), ‘Reducing the risk of wildfire’.

NAEI (2019), ‘UK emissions data selector’.

NAEI^{[footnote 45]} (2021), ‘Report: UK Greenhouse Gas Inventory, 1990 to 2019: Annual Report for submission under the Framework Convention on Climate Change’.

National Health Service, NHS (2016), ‘Adult Psychiatric Morbidity Survey: Survey of Mental Health and Wellbeing, England, 2014.’.

National Health Service, NHS (2021), ‘2019/20 National Cost Collection Data Publication’.

National Fire Chiefs Council, NFCC (2022), ‘The Economic and Social Value of UK Fire and Rescue Services’.

National World (2022), ‘Windsor Castle fire: what caused fire at Queen’s residence, how bad was the damage – cost of repairs explained’.

Nimblefins (2022), ‘Average Cost to Rebuild a House’.

OECD (2023), ‘Purchasing power parities (PPP)’.

Office for National Statistics, ONS (2013), ‘Number of hotels and hotel rooms in the UK’.

Office for National Statistics, ONS (2020a), ‘UK manufacturers’ sales by product’^{[footnote 46]}.

Office for National Statistics ONS (2020b), ‘Population estimates for the UK, England and Wales, Scotland and Northern Ireland: mid-2019’.

Office for National Statistics, ONS (2020c), ‘Nature of crime: criminal damage’.

Office for National Statistics (2021a), ‘Output in the construction industry’^{[footnote 47]}.

Office for National Statistics (2021b), ‘Business and individual attitudes towards the future of homeworking, UK: April to May 2021’.

Office for National Statistics (2021c), ‘National life tables: UK’.

Office for National Statistics (2021d), ‘House prices: how much does one square metre cost in your area?’.

Office for National Statistics (2021e), ‘Annual Survey of Hours and Earnings time series of selected estimates’.

Office for National Statistics, ONS (2021f), ‘UK Hotels in 2019 and 2020’.

Office for National Statistics, ONS (2022a), ‘Employment rate (aged 16-64, seasonally adjusted): %’.

Office for National Statistics, ONS (2022b) ‘UNEM03: Unemployment by previous industrial sector’.

Office for National Statistics, ONS (2022c), ‘Earnings and hours worked, all employees: ASHE Table 1’^{[footnote 48]}.

Office for National Statistics, ONS (2022d). ‘Earnings and hours worked, all employees: ASHE Table 1’^{[footnote 49]}.

Office for National Statistics, ONS (2022f), ‘EMP13: Employment by industry’.

Office for National Statistics, ONS (2022g), ‘Consumer price inflation tables’^{[footnote 50]}.

Office for National Statistics, ONS (2022h), ‘Is hybrid working here to stay?’^{[footnote 51]}.

Office for National Statistics, ONS (2023) ‘CPI ANNUAL RATE 00: ALL ITEMS 2015=100’.

Öster, C., Willebrand, M., Ekselius, L. (2011), ‘Health-Related Quality of Life 2 Years to 7 Years After Burn Injury’, The Journal of Trauma and Acute Care Surgery, 71(5): pp. 1435-1441.

Plimsoll (2021), ‘Fire Protection Industry’.

Price Waterhouse Cooper, PWC (2018), ‘Growing through uncertain times: UK hotels forecast 2019’.

Holavanahalli, R. K., Helm, P. A., Kowalske, K. J. (2010), ‘Long-Term Outcomes in Patients Surviving Large Burns: The Skin’, Journal of Burn Care & Research, 31(4): pp. 631-639.

Richardson, J. K. (1995), ‘Total cost of fire in Canada’, Fire Research News, 75: pp. 1-2.

Salomon, J. A., Haagsma, J. A., Davis, A., de Noordhout, C. M., Polinder, S., Havelaar, A. H., Cassini, A., Devleesschauwer, B., Kretzschmar, M., Speybroeck, N., Murray, C. J., & Vos, T. (2015). ‘Disability weights for the Global Burden of Disease 2013 study’, The Lancet Global Health, 3(11): e712–e723.

Schaenman, P., Stern, J., Bush, R. (1995) ‘Total cost of fire in Canada: an initial estimate’ National Research Council Canada.

The Big Red Guide, BRG (2009), ‘New Research Puts UK Fire Protection Market At £3 Billion’.

Vision Research Reports (2020), ‘Fire Protection Systems Market Size, Share, Trends, Growth, Production, Consumption, Revenue, Company Analysis and Forecast 2021-2028’.

Zhuang, J., Payyappalli, V. M., Behrendt, A., Lukasiewicz, K. (2017), ‘Total Cost of Fire in the United States’.

8. Glossary

Accidental Fire: A fire where the motive for the fire was presumed to be either accidental or not known (or unspecified)^{[footnote 52]}.

Arson: An arson offence is defined in the Criminal Damage Act of 1971 as ‘an act of attempting to destroy or damage property, and/or in doing so, to endanger life’.

Average cost: The total cost (anticipation cost plus consequence cost plus response cost) divided by the number of incidents.

Control staff: Uniformed staff working in FRS control centres to answer emergency calls and deal with mobilising, communications, and other related activities.

Deliberate fire: A fire where the motive for the fire was ‘thought to be’ or ‘suspected to be’ deliberate. This includes fires to an individual’s own property, others’ property or property of an unknown owner.

Dwelling fire: A fire in a property that is a place of residence; places occupied by households such as houses and flats, excluding hotels/hostels and residential institutions. Dwellings also include non-permanent structures used solely as a dwelling, such as houseboats and caravans.

Home Fire Safety Check: HFSCs are often targeted towards the more vulnerable members of the community, including older people and those with disabilities, and aim to offer bespoke fire safety advice based on their household and lifestyle which can mitigate the risk of fire. These visits can result in the installation of smoke alarms and safety recommendations such as moving or removing furniture which may be hazardous. Many FRSs now conduct HFSCs as part of a wider SWV, which can cover additional factors such as fall risk assessment, smoking cessation, cold homes and fuel poverty and a range of other health and community issues depending on local arrangements.

Marginal cost of Fire: The cost incurred following a fire, that would be directly impacted by changing fire numbers.

Other Building Fire: A fire in other residential or non-residential buildings. Other (institutional) residential buildings include properties such as hostels/hotels/B&Bs, nursing/care homes, student halls of residence etc. Non-residential buildings include properties such as offices, shops, factories, warehouses, restaurants, public buildings and religious buildings.

Other Outdoors Fire: A fire in either primary outdoor locations, or a fire in non-primary outdoor locations that have casualties or 5 or more pumping appliances attending. Outdoor primary locations include aircraft, boats, trains and outdoor structures such as post or telephone boxes, bridges and tunnels.

Primary Fire: A fire that meets at least one of the following conditions: any fire that occurred in a (non-derelict) building, vehicle or (some) outdoor structures any fire involving fatalities, casualties or rescues any fire attended by 5 or more pumping appliances.

QALY (Quality Adjusted Life Year): The monetary value of a life year, adjusted for health-related impacts on quality of life.

Road Vehicles Fire: A fire in a vehicle used for transportation on public roads, such as cars, vans, buses/coaches, motorcycles and lorries/HGVs. ‘Road vehicles’ does not include aircraft, boats or trains, which are categorised in ‘other outdoors’.

Secondary Fire: Small outdoor fires, not involving people or property. These vehicles include refuse fires, grassland fires and fires in derelict buildings or vehicles, unless these fires involved casualties or rescues, or 5 or more pumping appliances attended, in which case they become primary other outdoor fires.

Unit Cost of Fire: The change in total economic and social cost from one additional fire occurring. The unit cost includes consequence costs and response costs and excludes anticipation costs.

1. All costs in this report are converted into year ending March 2020 prices using the ONS GDP deflator or otherwise named conversions so they are consistent. As per HMT Green Book 2022 (HMT, 2022) guidance on government appraisal, future costs should be adjusted for inflation and discounted appropriately into single year prices (the base year). In this case, the base year chosen is year ending March 2020, so any costs falling after this year that were caused by fires that occurred in year ending March 2020 (such as physical and emotional harms from fatalities and long-term injuries), are discounted back to year ending March 2020 prices appropriately using a 1.5% or 3.5% discount rate. ↩

2. For example, expenditure in buildings and consumer goods to ensure materials are fires safe would be desirable and incurred even if the number of fires that occurred that year was zero, as in this case, the amount of expenditure drives the total number of fires, as opposed to the number of fires driving societal costs (such as in the consequence and response sections of the report). ↩

3. Including the DCLG cost of fire reports, and NPFA study in the USA. ↩

4. The Regulatory Reform (Fire Safety) Order 2005. As costs are based on impact in year ending March 2020, no costs from legislative changes such as the Fire Safety Regulations 2022 are considered in scope. ↩

5. Internal equivalents were considered, but due to differences in building regulations and building designs across countries, they were deemed to be less accurate proxies. International comparisons are considered as a sense check of the total costs in 3.1.5. ↩

6. Using same methodology as 3.1.1. The remaining cost of construction not spent on new builds is related to repairs and maintenance. £155.0 billion minus £101.4 billion is £53.6 billion. ↩

7. Page 4-24 to 4-25. ↩

8. Using a single percentage assumption should be heavily caveated as the cost of repairs and level of fire safety expenditure will vary largely by building type. It is also likely that in year ending March 2020, that some of the in-scope repairs and maintenance changes in buildings were due to remediation for fire safety reasons. This means that in year ending March 2020, the proportion of fire safety spend may have been higher than the “usual” average. As discussed at the start of this report, this spend is considered out of scope of this report, as it may skew these estimates. ↩

9. Some maintenance costs may be sunk in the case of fixed contracts with no break clauses; however these costs are considered to be low, and it is seen to be beyond the scope of this report to include or account for these costs. ↩

10. Department has since been renamed to the Department for Levelling Up, Housing and Communities. ↩

11. Multiplied by 87%, which is the best estimate for the proportion of the UK construction market which is England only. This figure differs slightly to the figure used in the DCLG reports because of an update to this conversion factor. ↩

12. Based on evidence that some aspects of the fire protection market have experienced significantly higher growth rates within the last 5 to 10 years, and are forecasted to potentially continue larger growth rates. ↩

13. The figures are multiplied by 88% (GDP conversion), then 68% (PPP conversion) then 17% (population conversion). Total multiplication of 10%. ↩

14. Manufacturing output estimated to be USD$2,193 billion in the US in 2017 (Macro Trends, 2022) (54/2,175 = 2.46%), AUD$113 billion in 2009 (1.6 / 113 = 1.42%) and CAD$123 billion in 1995 (2.0 / 123 = 1.52%). Converted into appropriate currencies using PPP data (OECD, 2023). ↩

15. 1.42% is used, rounded to 1.4% for simplicity in this report. ↩

16. Websites such as Register my appliance, GOV.UK and Electrical Safety First provide lists of goods subject to product recalls or product safety alerts. ↩

17. Economic and social cost of crime report covers England and Wales, so England only estimates by calculating figures by 95% (the proportion of the England and Wales population living in England). ↩

18. This does not mean to imply that the local FRS would not attend these incidents, it just indicates that any firefighting the fixed fire suppression system does, would have been completed by the local FRS in the absence of the fixed fire suppression system. ↩

19. In February 2023, the Department of Business, Energy and Industrial Strategy (BEIS) was disbanded, and the Department for Energy Security and Net Zero (DESNZ) and the Department for Business and Trade (DBT) were introduced in its place. ↩

20. The published BEIS value is £241 for 2020 in 2020 prices, which is converted to a year ending March 2020 value for this report. To do this, first the cost is divided by 1.015 in line with BEIS guidance that the cost increases by 1.5% annually. This gives a unit cost of £237 for 2019 in 2020 prices. Using the GDP deflator (HMT 2023), this cost is converted into 2019 prices (£224) and multiplied by 0.75 (£168). This is then added to the 2020 carbon value in 2020 price multiplied by 0.25 (£60) to give the year ending March 2020 carbon value in year ending March 2020 prices (£228). ↩

21. This uses the same rationale as the Home Office Economic and Social cost of crime report (HO, 2018: pg. 27-28). ↩

22. The Regulatory Reform (Fire Safety) Order 2005. ↩

23. ASHE ONS data shows that the median hourly gross wage is £13.71. This is uplifted by 18% to account for non-wage costs, giving a total hourly wage of £16.16. ↩

24. This includes the health, and firefighter injury sections of the report. ↩

25. With the exception of communal living properties which are classed as “other buildings” in the IRS but have the dwelling uprate factor applied. ↩

26. The 344m² figure was found by working backwards through their calculations of total damage area and total unemployment. ↩

27. Industrial premises, retail premises, Hotel, boarding houses hostels, food and drink premises, offices, and call centres, entertainment, culture and sport, and agricultural premises. ↩

28. It is estimated that there 154,000 fires were attended in England in year ending March 2020 (HO, 2023, FIRE0102), and 44,000 fires attended across Wales, Scotland, and Northern Ireland (NFCC and National data collections). 154/ (154+44) = 0.78. ↩

29. See Section 3.4 for the explanation of how the year ending March 2020 unit cost of carbon is calculated using the 2020 BEIS value for the unit cost of carbon. ↩

30. Published figure is £74,769 in 2022 prices. The GDP deflator (HMT, 2023) was used to adjust this figure to 2019 and 2020 prices (£66,900 and £71,100 respectively). These figures are then multiplied by 0.75 and 0.25 respectively, then summed, to give a year ending March 2020 price years total. ↩

31. There were 200,500 outdoor fires in year ending March 2006, 144,000 in year ending March 2009, and 87,600 in year ending March 2020. Therefore, the figures are multiplied by 0.44 and 0.61 respectively. The figures are then multiplied by 0.78 to account for differing fire numbers across England and the rest of the UK (see footnote 28). ↩

32. Vegetation fires are assumed to be equal to “other outdoors”. Vegetation fires include wildfires, prescribed forest fires, and open-field burnings related to agricultural practices, therefore being closely aligned to the Home Office definition of outdoor fires. However, it is possible that this figure is a slight underestimation compared to the Home Office definition of all outdoor fires, as it is unlikely all categories of other outdoor fires (such as smaller secondary fires, which do emit air pollutants) are included within this estimate. ↩

33. Calculated by multiplying the number of outdoor fires in the forestry commission definition of a wildfire incident in year ending March 2020 (26,000), by the average hectares of damage per wildfire (0.14). The latter is fully explained in 1.5.3. ↩

34. 13 calculated as 3,700 / 278 hectares. 61% is the proportion of damage from outdoor fires in the IRS which is caused by fires with a burn area of over 5,000m². ↩

35. Primary cost exceeds secondary as the majority of the damage from the largest incidents is from primary incidents. ↩

36. £500,000 loss to the local economy versus £1.8 million on suppression and restoration = 28%. ↩

37. The IRS does not include a date-time stamp for when an appliance returns to the station, so an assumption has had to be made. This may underestimate the real impact on FRSs, as they will likely use blue lights on the way to incidents, and therefore travel faster. Equally, on some occasions there will be no return time cost as the appliance will mobilise to another incident (which will be classed as drive time). ↩

38. 29 services responded to the crewing survey. ↩

39. As a sense check, the average when not weighting services (to prevent large services biasing the results) were also calculated. This was 83% (when excluding the Isles of Scilly as an outlier), so broadly aligned to the 72%. ↩

40. Full definition and further information on these staff available at HO (2022b). ↩

41. Excluding London due to London additional pay weighting. ↩

42. The cost of responding to non-fire incidents is expected to vary largely by non-fire incident type. Calculating the cost breakdown of different types of non-fire incidents is seen as beyond the scope of this report. ↩

43. There is no fire false alarm related anticipation spend accounted for in these calculations. ↩

44. 2019 data used as a proxy for year ending March 2020 figures. ↩

45. This included discussions with the NAEI team who complete these reports used. ↩

46. Divisional totals for 2019 used. This data is only available for calendar years, but this is seen as the best possible proxy for financial year data. ↩

47. Table 2a – volumes seasonally adjusted by sector. ↩

48. Table 1.5a, Year ending 2020 data used. Uplift factor sourced from https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/earningsandworkinghours/bulletins/indexoflabourcostsperhourilch/julytoseptember2020, with an average applied across the 4 quarters in year ending March 2020. ↩

49. Table 1.9a, Year ending 2020 data. ↩

50. Table 9, row 57-59. ↩

51. Data for 2020 used – sourced from link that says “Download the data for businesses using homeworking”. ↩

52. Where definitions relate to fire statistics, definitions are taken from HM Government’s published list of Fire Statistics Definitions (HO, 2022b). ↩