Research and analysis

Economic and social cost of fraud 2023 to 2024

Published 9 March 2026

Applies to England and Wales

Authors

Fraser Cornwell and Ross Watkins

Acknowledgements

This report was authored by Fraser Cornwell and Ross Watkins, with support from Marco Minasi-Smith.

The authors would like to give special thanks to all who provided insights to make this report possible. This included individuals across law enforcement, government 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 to Professor Mark Button of the University of Portsmouth, who peer reviewed the final report.

Executive summary

This report estimates the economic and social cost to society of fraud against individuals and businesses in England and Wales in the financial year ending (YE) March 2024. While previous estimates of the economic and social cost of fraud have only included fraud against individuals, this edition is the first iteration to include an estimate of the social and economic cost of fraud to businesses.

Following the established methodology of previous ‘Economic and Social Cost of Crime’ reports published by the Home Office, this report considers 3 main cost areas:

1. Costs in anticipation of fraud – incurred to protect from fraud prior to an incident of fraud.
2. Costs as a consequence of fraud – financial losses and emotional harms incurred by victims of fraud following an incident of fraud.
3. Costs in response to fraud – incurred to investigate, enforce and operate the legal system and other institutions in fraud cases.

The cost of fraud against the public sector is out of scope in this analysis. Fraud against the public sector is the responsibility of the Public Sector Fraud Authority (PSFA).

This report finds that the total economic and social cost of fraud against individuals and businesses in England and Wales was £14.4 billion in YE March 2024. This comprises of:

• costs in anticipation of fraud – £5.0 billion, 35% of total cost
• costs as a consequence of fraud – £8.0 billion, 55% of total cost
• cost in response of fraud – £1.4 billion, 10% of total cost

Of this total cost, fraud against individuals was estimated to be £9.2 billion and fraud against businesses was estimated to be £5.2 billion. Anticipation costs were estimated to be £5.0 billion – approximately 83% of these costs were incurred because of fraud against businesses. Consequence costs were estimated to be £8 billion – 90% of these costs were incurred because of fraud against individuals. Costs in response to fraud are estimated to be approximately £1.4 billion – 80% of these costs were incurred because of fraud against individuals. The costs in this report are all in YE March 2024 prices.

Table 1a: Economic and social cost of fraud - total cost

	Costs in anticipation (billions)	Costs as a consequence (billions)	Costs in response (billions)	Total costs (billions)
Total	£5.0	£8.0	£1.4	£14.4
Fraud against individuals	£0.9	£7.2	£1.1	£9.2
Fraud against businesses	£4.2	£0.8	£0.3	£5.2

Table 1b: Economic and social cost of fraud - unit cost

	Costs in anticipation	Costs as a consequence	Costs in response	Total costs
Fraud against individuals	£274	£2,255	£354	£2,884
Fraud against businesses	£1,145	£763	£262	£2,170
Total	£490	£1,885	£331	£2,707

Notes (Table 1a and 1b):

1. Figures may not sum due to rounding.

1. Introduction

1.1 Aim of analysis

The purpose of the ‘Economic and social cost of fraud’ report is to provide a robust estimate of the economic and social impact of fraud against both individuals and businesses in England and Wales in the YE March 2024. This report follows the costing framework of ‘The economic and social costs of crime, second edition’ (Heeks and others, 2018) to ensure consistency with other Home Office ‘cost of’ publications. This report builds upon previous estimates of the economic and social cost of fraud in ‘Fraud Strategy: stopping scams and protecting the public’ (Home Office, 2023), to include methodological refinements and updated insights, and expands to cover fraud against businesses.

1.2 What is included?

This report presents a government-led estimate of the economic and social cost of fraud affecting individuals and businesses across England and Wales. This report produces a conservative, credible, and robust baseline for understanding the scale and impact of fraud that is methodologically consistent with previous ‘cost of’ reports. To maintain reliability and consistency with other ‘cost of’ reports within government, the use of external sources has been limited — particularly where the methodology, origin or robustness of data cannot be verified. A conservative approach has been prioritised to ensure estimates are robust and not overestimated, including careful efforts to eliminate double counting across cost categories. Therefore, the resulting estimates can be considered a robust and credible baseline for the economic and social cost of fraud on society for use in policy development and discussion.

This report takes the definition of fraud from the Crime Survey of England and Wales (CSEW) (ONS, 2020) of ‘a person dishonestly and deliberately deceiving a victim for personal gain of property or money, or causing loss or risk of loss to another’, aligning to the Fraud Act 2006. As such, cybercrime and ransomware attacks are out of scope of this report, as these crimes come under the Computer Misuse Act 1990. This analysis breaks down the cost of fraud into fraud that has been committed against individuals and businesses. It is out of the scope of the report to split the cost of fraud into different types of fraud due to data limitations.

The cost of fraud against the public sector is out of scope in this analysis. Fraud against the public sector is the responsibility of the Public Sector Fraud Authority (PSFA). A recent National Audit Office (NAO) overview of the impact of fraud and error on public funds was estimated to cost the taxpayer around £55 billion to £81 billion in YE March 2024 (NAO, 2024).

This report follows previous ‘cost of’ reports’ structure and divides costs into 3 main areas:

1. Costs in anticipation of fraud – incurred to protect from fraud prior to an incident of fraud.
2. Costs as a consequence of fraud – financial losses and emotional harms incurred by victims of fraud following an incident of fraud.
3. Costs in response of fraud – incurred to investigate, enforce and operate the legal system and other institutions in fraud cases.

This report aims to monetise as much of the cost of fraud to society as possible. However, there are various potential costs of fraud to society not monetised in this analysis. These include:

Very low probability, high-cost frauds: The Economic Crime Survey (ECS) 2024, may have not captured rare, high-impact fraud incidents affecting a very small proportion of businesses.

‘Hidden’ incidents: The ECS 2024 only reports economic crimes that businesses could detect or were willing to disclose. Since detecting fraud is difficult, the actual number of incidents and affected businesses is likely higher than reported. Estimating undetected or undisclosed crimes was outside the scope of the ECS, therefore was not included in this analysis. It is also possible that this is the case for fraud against individuals whereby individuals are not aware they have fallen victim to fraud.

Emotional harms to employees: The analysis focuses on the financial impact of fraud on businesses, excluding emotional harm to employees. Emotional harm may vary based on the employee’s role and the size of the fraud, but due to lack of evidence, it was not included in this report. Therefore, this is likely underestimating societal costs.

Opportunity cost of losses: Only the direct financial loss is measured in this report. There are other costs because of the losses incurred by individuals and businesses from incidents of fraud, such as missed investment opportunities or further impacts on consumption levels of individuals. However, this has not been considered due to a lack of evidence.

Reimbursement of losses: To avoid double counting, reimbursement costs were not included in this analysis. This is an economic transfer between victims and banks. If banks cannot recover funds from fraudsters, the cost falls on the finance industry and the economy. There is also a lack of data on partial reimbursements and their average amounts. Future work will explore the size of recovered funds and the impact of reimbursement levels on the cost of fraud to society.

1.2.1 Key assumptions

There are various key assumptions that guide the scope and methodology of the report, influencing the estimated costs of fraud against individuals and businesses.

1. Finance industry data: insights into the cost of fraud to the finance industry can be considered as a conservative industry estimate and the true cost may in fact be higher. This estimate comes with several assumptions:

   • not all banks were able to provide data within the deadline
   • data from smaller Payment Service Providers (PSPs) were extrapolated to represent the whole industry
   • costs in anticipation and response to fraud were assumed to be split 50/50 where exact data was unavailable
   • data provided on a ‘best endeavours’ basis, with some figures estimated
   • financial industry costs cover Scotland and NI alongside England and Wales as data on financial industry costs cannot not be to split by country

2. Criminal Justice System (CJS) Costs: Ministry of Justice data does not differentiate by victim type, so the analysis assumes total response costs include both individuals and businesses. Total costs are apportioned to each cohort in accordance with the Action Fraud reporting split (90% individuals, 10% businesses). Therefore, assuming conviction rates are equal for both fraud against individuals and businesses.

Assumptions for individuals

• age: Individuals aged 16 and under are out of scope of this analysis; CSEW data does not collect data for this cohort and there was a lack of evidence on victimisation rates of this cohort to accurately extrapolate from CSEW; this means the total cost is likely an underestimate
• protection costs: only financial industry costs are considered, assuming individuals do not purchase anti-fraud capabilities themselves
• emotional harms: it was only possible to monetise fear, depression and anxiety, likely underestimating the total emotional harm
• quality-adjusted life year (QALY) loss: based on 2013 data, assuming constant prices over time
• police time: uses results from the 2022 to 2023 Police Activity Survey, likely underestimating opportunity costs as it excludes several forces, including the Metropolitan Police Force
• fraud enforcement costs: includes spending from the National Crime Agency, City of London Police and Serious Fraud Office
• CJS costs: uses current cost of crime modelling, including prison and probation costs for those on remand; youth offenders are not considered in this model due data limitations on the level of fraud-specific offenders aged 10 to 17
• civil justice costs: there are other avenues that victims of fraud can seek justice through, such as private prosecutions, civil justice and regulatory justice; these are not included in this analysis due to a lack of evidence of the volumes of cases which are fraud specific

Assumptions for businesses (ECS 2024)

• Economic Crime Survey (ECS) 2024: only includes businesses with employees - , likely underestimating costs as businesses with zero employees are excluded
• geographical scope: this analysis takes insights from ECS for UK businesses and scales them to England and Wales only; this is to align with individual modelling, which uses CSEW, and existing cost of crime analysis
• incident calculation: uses the prevalence rate of fraud and the median number of frauds per business to estimate incidents
• cost calculation: uses the mid-point methodology for ECS loss bands, assuming equal distribution of losses within each band
• no overlap in defensive expenditure: assumes the defensive expenditure captured in ECS does not overlap with estimates provided by the financial industry; costs provided by the finance industry relate their role in protecting customers, be it individuals or business, rather than protection against being defrauded themselves

1.2.2 Limitations and future analysis

By its nature, fraud is hidden, perpetrators go to great lengths to conceal their actions from law enforcement, there is often a long time without detection, and many victims are also reluctant to report fraud. Whether this is due to perceived reputational impact to businesses or a reluctance from individual victims to avoid reliving psychological harm felt at the point of victimisation. As a result, many incidents of fraud go unreported, undetected and are not investigated.

Therefore, estimating the true cost of fraud to society is challenging. Where available this report uses long-standing, nation-wide surveys such as CSEW and ECS, alongside subject matter expertise and bespoke data to calculate as accurate and robust estimate as possible.

This analysis is limited to aspects of fraud where data availability allows for the quantification of estimated impacts. As a result, this estimate is likely to underestimate the true cost of fraud to society. Future iterations of the economic and social cost of fraud will aim to improve insights, expanding to include:

• costs incurred by individuals in anticipation of fraud, especially the inclusion of spending by individuals into anti-fraud capabilities
• expanding the monetisation of psychological harms felt by victims of fraud beyond fear, depression and anxiety
• explore the possibility of expanding ECS to include all businesses, not just businesses with employees

2. Cost of fraud against individuals

2.1 Scope of the section

This section outlines the economic and social cost of fraud against individuals. This analysis follows the approach of existing ‘cost of crime’ analysis to split costs into 3 main categories. Such that:

Cost of fraud = Costs in anticipation + Costs as a consequence + Costs in response

Table 2: Components of the cost of fraud against individuals

Area of cost	Total cost (billions)
Costs in anticipation of fraud	£0.9
Defensive expenditure - The costs incurred in activities to prevent fraud.	£0.9
Insurance administration - It is assumed that individuals do not buy insurance with fraud coverage due to a lack of evidence on the levels of policies which might cover fraud-related incidents.	£0.0
Costs as a consequence of fraud	£7.2
Financial loss - The value of money taken by, or given to, fraudsters as a result of attempted fraud.	£3.2
Emotional harm - The reduction in the quality of life of the victim from emotional harm suffered as a result of fraud.	£1.9
Lost output - This estimates the lost productivity from time off work and reduced productivity whilst experiencing the psychological effects from fraud.	£1.1
Health service costs - The cost of dealing with emotional harms of crime, including counselling costs.	£0.9
Victim support - Estimates the cost of support provided to victims of fraud by the National Economic Crime Victim Care Unit (NECVCU).	£0.0
Costs in response to fraud	£1.1
Police and law enforcement costs - The opportunity cost of police officers’ investigation against fraud. This captures the value of police time that is taken up by investigating fraud, rather than by engaging in other crime types or other non-crime activity.	£0.2
CJS costs - The CJS is a set of agencies and processes established to control crime and impose penalties on those who break the law. The costs include those for the Crown Prosecution Service (CPS), court costs, Legal Aid, prison costs, remand and probation.	£0.2
Finance industry costs - Provided by the finance industry into the response costs as a result of fraud, such as running operations and support teams, strategy and industry engagement.	£0.7
Total cost	£9.2

Notes:

1. Figures may not sum due to rounding. 2, Victim support: NECVCU costs are estimated to be £3.2million.

2.2 Costs in anticipation of fraud

Anticipation costs are incurred as an attempt to detect and block fraud. An example of this would be spending by financial institutions on anti-fraud operations. Total anticipation costs for fraud against individuals are estimated to be £871 million, and this section summarises the costs.

This report assumes that only businesses bear the costs of defence and insurance administration. Individuals do not buy insurance or anti-fraud measures themselves; instead, they rely on their banks and the businesses they interact with for protection. This is likely not the case, and some individuals will take some steps to defend themselves from fraud which may cost individuals, therefore resulting in an underestimate of the total defence costs. Products are coming to the market which individuals can purchase that aim to protect from fraud, especially within the identity protection space. This was not included in this report due to a lack of evidence of the size of this market and how it relates specifically to fraud prevention only.

Whilst individuals typically do not take out insurance specifically against fraud, many insurance policies such as home insurance could include coverage for fraud-related incidents. This is not included in this analysis, instead it is assumed that individuals do not buy insurance with fraud coverage due to a lack of evidence on the levels of policies which might cover fraud-related incidents.

The financial industry estimates that £970 million (YE March 2024 prices) is spent to prevent fraud. Whilst this cost is borne by a business, it is allocated to individuals and businesses in accordance with Action Fraud reporting splits by victim type. This allows for a calculation of anticipation costs for both individuals and businesses, given the assumption that individual victims do not incur anticipation costs, but the financial industry does regardless of victim type. Consequently, this analysis estimates that £871 million of this amount is spent to prevent fraud against individuals.

This includes costs incurred for prevention activities such as:

• resource costs (fraud operations and strategy teams, a proportion of branch headcount and call centre costs)
• prevention tools, technology and systems
• education and awareness for colleagues and consumers
• industry memberships such as CIFAS, UK Finance, Stop Scams
• strategic change costs

Table 3: Costs in anticipation of fraud

Cost type	Total cost (Millions)	Unit cost
Defence costs	£871	£274
Insurance administration costs	£0	£0
Total anticipation costs	£871	£274

2.3 Costs as a consequence of fraud

Consequence costs are incurred because of fraud victimisation. For example, the financial losses and emotional harms suffered. Total consequence costs of fraud against individuals are estimated to be £7.2 billion, and this section summarises the costs.

2.3.1 Financial loss

Individual financial loss is calculated using CSEW loss data. The CSEW separates losses into bands based on the amount lost, for example, £50 to £99. By multiplying the proportion of incidents that fall into each band by the total volume of fraud, the number of incidents in each band is estimated. These are then multiplied by the mid-point of their respective bands (for example, £74.50 in the £50 to £99 band), and summed, to estimate the total loss. For the highest band (£40,000+) a loss of £60,000 was selected as this was in line with the increase in loss per incident between bands. This approach estimates the total expected loss of fraud against individuals to be around £3.2 billion. To calculate the unit cost of financial loss, the total loss is divided by the number of loss-incurring fraud incidents. The result is an average loss to individuals in incidents of fraud of £1,008.

While financial loss from fraud against individuals is likely to be an underestimate, this approach accounts for the wide variation in fraud losses by using a robust estimate that reflects how case volumes differ by loss size, avoiding reporting bias. Action Fraud data includes reported losses from both businesses and individuals. However, this data is highly skewed as more cases report no losses, while few cases report losses into the millions. Therefore, mean and median losses from Action Fraud are not appropriate to estimate the financial loss of a typical fraud case. Fraud incidents are also under-reported, data from ONS estimates suggest that only 14% of fraud is reported to the police or Action Fraud (see ONS, 2024).

2.3.2 Emotional harm

Fraud does not cause direct physical harm to victims but can inflict significant psychological harm. Due to data constraints, this analysis focuses on the first order psychological effects of falling victim to fraud. The CSEW publishes information on the number of people who experience specific emotional harms as victims of fraud. The model uses a QALY approach, as found in the ‘Economic and social cost of crime’ (ESCC). This is applied to monetise the emotional impacts associated with victims of fraud, which are anxiety, depression and fear. There are many other types of psychological harm which could be experienced by victims of fraud. However, fear, depression and anxiety were the only harms available to be monetised and considered in this analysis.

For each injury (i):

LIKE_i * REDUCEQL_i * DUR_i * VOLY = Average physical and emotional cost_i

Where:

LIKE – Likelihood of sustaining a specific physical and/or emotional injury.

REDUCEQL – Percentage reduction in quality of life for that physical/emotional injury, informed by Solomon et al (2015).

DUR – Duration of the injury as a fraction of a total year, informed by Dolan et al (2005).

VOLY – Value of a year at full health, assessed by DHSC to be £70,000 in 2022 prices (HM Treasury, 2022).

2.3.3 Lost output

There may be a reduction in economic activity when an individual becomes a victim of a crime. This section estimates the cost of lost productivity for fraud victims who may proceed to take time off work and potentially be less productive for some time after. The following costs are included:

1. Time taken off work due to the fraud incident.
2. Reduced productivity due to the emotional harms sustained from the incident.

Our approach follows the same methodology as set out in the ESCC report. This is based on CSEW respondents who report the amount of time taken off work, and QALY loss as a proxy for the extent of their reduced productivity.

For each injury (i):

LIKE_i * LOSTHRS_i * HRLYW_i * EMPRATE_i = Average lost output cost_i

Where:

LIKE – Likelihood of sustaining a specific physical and/or emotional injury.

LOSTHRS – The lost productive hours as function of taking time off work and reduced productivity per hour once the victim has returned to work.

HRLYW – The mean hourly wage according to ONS data, including an uplift to account for non-wage costs.

EMPRATE – Employment rate of victims of fraud, from CSEW, to account for the lower employment rate for victims of crime.

2.3.4 Health services costs

Health service costs captures the costs to the NHS and other healthcare providers of responding to the physical and emotional harms of crime. This analysis exclusively uses NHS costs. If an individual were to use private healthcare, then the costs in this report are likely to be underestimated. If people are injured as a result of crime, the resources used to treat them could be used in alternative activities. Therefore, these costs are an opportunity cost (social cost) to society.

In line with ESCC methodology, the unit costs of healthcare activities are derived from Curtis and Burns (2020) and NHS England (2021). This analysis monetises exclusively direct psychological harm of falling victim to fraud. Therefore, associated average costs come from the number of counselling hours required, multiplied by the hourly cost of care. The number of counselling hours required for each harm type is informed by Dubourg and others (2005). This figure is calculated for fear, depression and anxiety. This methodology follows that used in the previous ESCC, as there is no evidence specifically on the volume of fraud victims who seek counselling following an incident of fraud.

To begin to calculate the unit cost of fraud, the average cost for each harm is multiplied by the likelihood of a victim suffering that harm. Each expected harm value is then added together to calculate the unit cost of fraud.

Healthcare costs = (Fear unit cost * likelihood of victim fear) + (Depression unit cost * likelihood of victim depression) + (Anxiety unit cost * likelihood of victim anxiety)

Victim Services

The National Economic Crime Victim Care Unit (NECVCU) provides support to both vulnerable and non-vulnerable individuals who have fallen victim to fraud. This includes tailored support and advice to help victims emotionally and practically from their experiences, with an aim to prevent re-victimisation and build consumer confidence. Separate to health services costs calculated above, the NECVCU have provided an estimate of the cost to run the unit at approximately £3.2 million.

Table 4: Costs as a consequence of fraud

Type of cost	Total cost (millions)	Unit cost
Financial loss	£3,203	£1,008
Emotional harms	£1,946	£613
Lost output	£1,128	£355
Health services	£887	£279
Victim services	£3	£1
Total consequence costs	£7,169	£2,256

Notes:

1. Figures may not sum due to rounding.

2.4 Costs in response to fraud

Response costs to fraud are incurred when action is taken following an incident of fraud. Total response costs of fraud against individuals are estimated to be £1.1 billion, and this section summarises the costs.

2.4.1 Policing and enforcement costs

This analysis uses a different method to calculate the opportunity cost of police time spent on fraud offences than the method used for the ESCC. Previous methods used data from the YE March 2007, but fraud has changed significantly since then due to increased internet and smartphone usage. This method uses results from the 2023 Police Activity Survey (PAS), which collected data from 35 police forces in England and Wales (Home Office, 2025a). This survey was completed by all police officers and staff in each participating force over a 7-day survey period. The survey was not mandatory for police forces and personnel; however, over the course of the survey, the Home Office collected over 160,000 responses with over 6.1 million hours of activities recorded across 35 forces. Respondents detail how they spent their time on shift across a range of police activities.

PAS activity time is split into 3 categories:

1. Activities linked to specific crime incidents.
2. Activities linked to specific Public Safety and Welfare (PSW) incidents.
3. General crime/PSW activities that are not tied to a specific incident.

Crime response workforce costs are calculated for each crime type by combining the costs of specific crime incident activities with the share of general crime/PSW activity costs that are attributable to crime response. These general activity crime response costs are then allocated to crime types according to each crime type’s share of overall crime workforce costs.

Insights from the National Crime Agency, City of London Police and Serious Fraud Office (SFO) have also been included in this analysis following engagement from subject matter experts in each institution. SFO has both investigative and prosecutorial responsibilities. Both have been considered and accounted for in this analysis. For simplicity, SFO costs are included in enforcement costs only.

2.4.2 CJS Costs

Crown Prosecution Service costs
Total CPS expenditure is estimated at £863 million for the YE March 2024. This is apportioned by offence type and court level using operational data and offence volumes to estimate the share of CPS resources used in fraud cases.

Magistrates’ court costs
For Magistrates’ court costs, Ministry of Justice (MoJ) data is used to find the number of fraud offences and split by plea type (see MoJ, 2024). This analysis assumes no change in average hearing times in Magistrates’ courts since ESCC due to lack of data. Multiplying the cases by the time taken for each plea type calculates the total court time, assuming a sitting day is 5 hours. The total court time is then multiplied by the cost per sitting day, adjusted to YE March 2024 prices.

Crown court costs
Crown court costs are calculated similarly, using MoJ data to find the number of fraud offences and split by plea type. Crown court hearing times are calculated using the average waiting hearing time tool for fraud and all offences (see MoJ, 2025). Multiplying the cases by the time taken for each plea type calculates the total court time, assuming a sitting day is 3.3 hours. The total court time is then multiplied by the cost per sitting day, adjusted to YE March 2024 prices.

Jury service
Jury service is an opportunity cost to society, as members of the public who are chosen for jury service cannot engage in work for the duration of the case. Jury service costs are calculated by taking the average hearing time for fraud offences in Crown Courts, multiplying it by the number of hearings, and dividing by 3.3 hours for a sitting day. This is then multiplied by 12 jurors and split the total sitting days into employed and unemployed. For employed jurors, we calculate the opportunity cost based on the mean wage.

Legal aid
MoJ has provided data on the total costs of each category of legal aid and the number of cases where it was required at Magistrates’ and Crown Courts. These totals are split between representation at Magistrates’ Court and advocate graduated fees or litigator graduated fees at the Crown Courts. The cost of advice given at police stations to those receiving legal aid is also included in the total legal aid costs. These inputs are used to calculate an average cost of legal aid for fraud offences.

Non-legal aid
Non-legal aid defence costs are estimated by scaling up legal aid costs using a ratio for private legal assistance. This is adjusted for self-representation and combines non-legal aid costs for both courts.

Prison service costs
To estimate prison costs for fraud offences in the YE March 2024, we use a method to avoid double-counting past crimes. The number of people sentenced to prison for fraud in the YE March 2024 is multiplied by the average sentence length and the cost per prisoner per year of £53,801 (see MoJ and HMPPS, 2025). Sentences under 7 years are assumed to serve 50% of their sentence in prison (see Sentencing Council, 2026).

In the YE March 2024, 983 people were sentenced to prison for fraud. The average sentence length was 22.4 months, which becomes 11.2 months with the above 50% assumption. Total prison service cost is then calculated by multiplying the number of people sentenced to immediate custody by the average sentence length and the cost per prisoner per year.

Prison costs also include those held on remand who do not serve time after sentencing. Those who serve time or get community or suspended sentences are excluded to avoid double counting. Remand costs are calculated similarly, assuming the cost is the same as prison.

Probation costs
Probation costs are calculated by finding the number of suspended or community sentences and multiplying by the marginal cost of probation, adjusted to YE March 2024 prices. The cost of defendants held on remand before sentencing is included and assumes that this cost is the same as keeping someone in prison.

Licence costs are then calculated to account for the remainder of the sentence length of offenders who serve prison time. This analysis assumes that everyone sentenced to prison serves half their sentence in prison and half on licence. Costs of those on licence are calculated similarly to prison costs, which is the number of fraud offenders sentenced to custody multiplied by the marginal cost of licence per person per year multiplied by the length of time on licence, which is calculated using the average sentence length minus the time in prison.

2.4.3 Financial industry costs

Data from the financial industry estimates that costs in response to fraud total £770 million (YE March 2024 prices). This excludes reimbursement to avoid double-counting of economic costs. Using the total number of reported frauds against individuals and businesses from Action Fraud data to calculate a unit cost in response to fraud, this analysis estimates that £692 million of this total is spent as a response of fraud against individuals.

Such costs include:

• resource costs for all post-fraud touchpoints (for instance, operations teams, call centres, branch support)
• strategy costs, such as the implementation of risk-based payment delays, APP Reimbursement Requirement and the Payment Systems Regulator’s (PSR) Measure 1 data reporting
• industry engagement in response measures

2.4.4 Calculating CJS cost of fraud against businesses

Ministry of justice data does not split by type of victim therefore this analysis assumes that the total response cost of fraud includes both frauds against individuals and businesses. Therefore, total response costs have been apportioned to each cohort in accordance with the latest breakdown in Action Fraud data. Approximately, 90% of action fraud reports are against individuals with 10% reported to be against businesses. As a result, 90% of total response costs are estimated to be incurred through frauds against individuals and 10% for frauds against businesses.

Table 5: Costs in response to fraud against individuals

Type of cost	Total cost (Millions)	Unit cost
Finance industry response costs	£692	£218
Criminal justice system costs	£243	£77
Law enforcement costs	£186	£59
Total response cost	£1,125	£354

Notes:

1. Figures may not sum due to rounding.

3. Cost of fraud against businesses

3.1 Scope of the section

This section outlines the economic and social cost of fraud against businesses. This analysis follows the approach of existing ‘cost of crime’ analysis to split costs into 3 main categories. Such that:

Cost of fraud = Costs in anticipation + Costs as a consequence + Costs in response

Table 6: Components of the cost of fraud against businesses

Area of cost	Total cost (billions)
Costs in anticipation of fraud	£4.2
Defensive expenditure - Costs incurred in activities to prevent fraud such as staff training, spending on software and risk assessments. (Home Office, 2024)	£3.7
Insurance administration - Costs incurred through the administration of insurance premiums to cover fraud incidents. (Home Office, 2024)	£0.5
Costs as a consequence of fraud	£0.8
Financial loss - The value of money taken by or given to fraudsters as a result of attempted fraud. (Home Office, 2024)	£0.5
Lost output - This estimates the lost productivity incurred in the aftermath of fraud attempts. Including staff time to deal with fraud, indirect costs (such as share value, customer complaints or loss of investors etc). (Home Office, 2024)	£0.3
Costs in response to fraud	£0.3
Police enforcement costs - The opportunity cost of police officers’ investigation with fraud. This captures the value of police time that is taken up by investigating fraud, rather than engaging in other crime types or other non-crime activity.	£0.02
Criminal Justice System (CJS) costs - The CJS is a set of agencies and processes established to control crime and impose penalties on those who break the law. The costs include those for the Crown Prosecution Service (CPS), courts costs, Legal Aid, prison costs, remand, and probation.	£0.03
Private investigation costs - The estimated spending on employees monitoring or investigating fraud. (Home Office, 2024)	£0.1
Finance industry costs - Provided by the finance industry into the response costs as a result of fraud such as running operations and support teams, strategy, and industry engagement.	£0.08
Total costs	£5.2

Notes:

1. Figures may not sum due to rounding.

3.2 Data quality

The majority of this estimate of the cost of fraud against businesses is informed by the Economic Crime Survey 2024 (Home Office, 2024). The Home Office commissioned Ipsos, supported by Professor Michael Levi from Cardiff University, to design and undertake a new study of the prevalence and impact of economic crime among UK businesses with employees.

This study aimed to estimate the prevalence and scale of fraud, corruption and money laundering incidents across UK businesses. In the fraud space specifically, the ECS measures the economic cost of fraud to businesses experiencing fraud incidents, alongside quantifying and understanding:

• the impact of fraud on businesses
• the response of businesses to fraud incidents
• the processes businesses have in place to minimise or deal with fraud

The survey, designed and conducted by Ipsos, involved 3,477 UK businesses with employees. Fieldwork, incorporating a pilot phase, took place between February and August 2024. The data has been weighted by size and sector to represent the UK business population. Various insights were collected on the costs and spending associated with business fraud. These measures intended to capture cost and spending data comprehensively, both by splitting them across multiple questions capturing specific aspects, and by providing flexibility in the way businesses could respond. The survey was designed to be representative of the overall business population, so it is feasible to extrapolate from the cost and spending statistics for the average business facing fraud, to produce overall business fraud cost and spending estimates for the entire UK economy.

3.2.1 Key limitations to ECS

There are some key limitations to the Economic Crime Survey which are important to note:

Zero employee businesses are excluded: according to the 2024 business population estimates (DBT, 2024), there are approximately 4.07 million 0-employee businesses, compared with approximately 1.43 million businesses with employees. Although 0-employee businesses are expected to have lower fraud-related costs on average (matching the broad pattern seen in the survey data, where larger businesses have higher costs), the cumulative impact of their exclusion could be substantial.

‘Hidden’ incidents not included: the ECS is only able to measure incidents that businesses were able to detect and identify or were willing to disclose. Specifically, challenges with the detection and identification of economic crime were present across all 3 crime types, meaning that the results showing the number of fraud incidents experienced, and the number of businesses affected, are likely to be underestimates. Estimating ‘hidden’ economic crime which businesses are not aware of is very challenging and was not within the scope of the ECS, therefore is not included in this analysis.

Cost and spending data being subject to higher margins of error: these estimates are not based on the full sample. Instead, they are based only on the subgroup of businesses in the sample that experienced any fraud and were able to provide an associated cost estimate. This is 533 respondents out of the total 3,477. As such, the margins of error, using a confidence interval of 95%, for these extrapolated cost estimates will be higher than for the full sample.

Undervaluing of staff time: the survey attempts to capture the indirect staff time cost for dealing with fraud. Previous UK government research on measuring the staff cost when dealing with cyber security breaches has found that businesses find it difficult to accurately measure these costs, as not all require staff to log their time (see DSIT, 2024) – it can be hard to allocate part of a wage to fraud if staff work on multiple tasks. However, despite following best practices, the survey may underestimate these costs due to the difficulty in measuring them.

The survey sample missing rare but high-impact incidents: it is possible that a very few businesses have very costly frauds, which have an inordinate effect on the total economic cost of fraud across all businesses – any survey sample would be unlikely to capture these kinds of rare incidents, meaning the mean (average) result recorded in the survey would be lower than the true mean in the population.

3.2.2 Mid-point methodology

Banded responses were given to the cost questions in ECS 2024, allowing for expected value to be calculated by multiplying the proportion of weighted responses by the mid-point of each respective band (for example, £15,000 in the £10,000 to £20,000 band). This method ensures that some of this underestimate of the survey not capturing rare but high-impact events is negated by inflating the unit costs relative to the mean responses recorded in the survey. Unit costs are calculated by dividing the total cost of each component by the median number of fraud incidents experienced by businesses in England and Wales.

3.3 Costs in anticipation of fraud

Anticipation costs are incurred as an attempt to detect and block fraud. For example, spending by financial institutions on anti-fraud operations. Total anticipation costs of fraud against businesses are estimated to be £4.2 billion, and this section summarises the costs.

3.3.1 Financial industry costs

Data from the financial industry estimates that costs in anticipation of fraud total £970 million (YE March 2024 prices). Such costs can be considered as a conservative industry estimate and the true cost may in fact be higher. This cost is apportioned to individuals and businesses in accordance with Action Fraud reporting splits by victim type. This analysis estimates that £98 million of this total is spent in anticipation of fraud against businesses.

This includes costs incurred for prevention activities such as:

• colleague costs (fraud operations and strategy teams, a proportion of branch headcount and call centre costs)
• prevention tools; technology and systems
• education and awareness for colleagues and consumers
• industry memberships such as CIFAS, UK Finance, Stop Scams
• strategic change costs

3.3.2 Defensive expenditure

Respondents in the ECS were asked questions on their activities and spending in anticipation of potential fraud. This analysis estimates that businesses in England and Wales with employees spent approximately £3.6 billion in the last 12 months in defence against fraud. This includes insight around spending on:

• staff costs to monitor and investigate fraud risks
• fraud training or awareness raising activities
• digital software to prevent or detect fraud
• fraud risk assessments

This is a significant proportion of the cost of fraud against businesses at approximately 68%. Showing the impact of fraud on business through the fact that businesses are compelled to spend such amounts in anticipation of fraud to either stop frauds occurring altogether or limit the impact of fraud once they do occur. The potential losses without preventative measures are unknown. The cost of optimal prevention might be higher than actual fraud losses incurred, but without these measures, fraud costs could be so high that the spending on preventive measures are justified.

Defensive expenditure also represents a significant opportunity cost to business, a counterfactual world in which fraud did not occur would free up almost £3.6 billion in extra funding which could be reinvested in productive capacity.

Monitoring risks
Businesses spent an estimated £1.3 billion in the last year on staff time to monitor and investigate fraud risks, according to ECS 2024. This includes any spending on employees whose job involves fraud monitoring or investigation. If employees only spent part of their time on fraud, respondents estimated the amount based on the proportion of time spent.

Monitoring fraud incurs an opportunity cost, as staff time could have been used for other productive activities. ECS respondents detailed the amount spent on employees for fraud monitoring and investigation, including staff wages.

Training
Approximately £1.2 billion is estimated to be spent by businesses on fraud training or awareness-raising activities for staff or customers in the last 12 months. This includes:

• spending on staff paid for their time to develop or deliver training
• spending on staff time to attend training
• external payments made to training providers

Software to prevent and detect fraud
Approximately £800 million is estimated to be spent by businesses on any digital software to prevent or detect fraud in the last 12 months. This includes any installation or licensing costs.

Risk assessments
Approximately £340 million is estimated to be spent by businesses to undertake any fraud risk assessments in the last 12 months. This includes:

• spending on staff time to run any risk assessments
• any payments made to third parties to run risk assessments

3.3.3 Insurance administration costs

Businesses with employees in England and Wales spent about £5.8 billion on insurance policies covering fraud in the last 12 months, according to ECS data. Insurance premiums and pay-outs are economic transfers and do not represent an overall cost to society, as they are an economic transfer between individuals and businesses. The actual cost comes from the resources used by insurance firms to handle fraud-related claims, which could have been used for other productive activities.

The opportunity cost arises from the time and resources insurers spend on issuing fraud protection contracts and processing claims. This is a cost to society because these resources could have been used elsewhere. Using data from the Association of British Insurers (ABI), as used in previous ‘cost of’ reports, this report estimates that around £495 million was spent on insurance administration for fraud policies in the YE March 2024.

Table 7: Costs in anticipation of fraud against businesses

Type of cost	Total cost (millions)	Unit cost
Defence costs	£3,582	£924
Insurance administration costs	£495	£128
Finance industry costs	£98	£94
Total anticipation costs	£4,175	£1,145

Notes:

1. Figures may not sum due to rounding.

3.4 Costs as a consequence of fraud

Consequence costs are incurred because of fraud victimisation. For example, the financial loss and lost output of businesses. Total consequence costs of fraud against businesses are estimated to be £798 million, and this section summarises the costs.

3.4.1 Financial loss

Financial Loss can be interpreted as the amount of money taken from businesses by fraudsters as a direct result of fraud incidents. This analysis, using the results of ECS 2024, estimates that £507 million was lost to fraudsters by businesses in England and Wales with employees in the last 12 months. ECS asked respondents about the total external payments made to fraudsters in the last 12 months, therefore the unit cost per incident is calculated as the loss per business divided by the median number of fraud incidents.

This is highly likely to be an underestimate. ECS 2024, like any survey sample, has a low probability of capturing the rare but high-impact incidents. Such incidents can have an inordinate effect on the total costs of fraud to businesses and as a result, costs recorded in the survey will be lower than the true cost in the population. This survey is also only able to measure incidents that businesses were able to detect and identify or were willing to disclose, meaning that results are likely to be underestimates.

3.4.2 Lost output

Businesses were asked to provide insight on medium-term costs as a consequence of fraud. This includes questions around their spending on staff time dealing with fraud and other indirect costs in dealing with fraud.

This analysis combines responses from each to provide an estimate for the value of lost output as a consequence of fraud. Costs incurred by businesses as a consequence of fraud outside of direct losses are an opportunity cost. Staff activities and time dealing with fraud and other indirect costs incurred such as loss of intellectual property, would not be incurred in a world without fraud. Therefore, time and resource could have been otherwise engaged in other productive activities.

This analysis, using the midpoint method with ECS 2024 data, estimates that businesses with employees in England and Wales spend approximately £291 million in the last 12 months as a consequence of fraud. This breaks down in the following way:

• £232 million in staff costs dealing with fraud
• £59 million in other indirect costs in dealing with fraud

Staff costs
The value of staff time in dealing with frauds in the last 12 months, was £232 million. This includes how much staff would have got paid for this time, even if dealing with these issues are part of any staff members job role. This allows the calculation of the opportunity cost of staff time to deal with fraud incidents.

This is potentially an underestimate of the true cost to businesses in staff time to deal with fraud incidents. Previous UK government research on measuring the staff cost when dealing with cybersecurity breaches has shown that businesses find it difficult to account for these costs accurately (see DSIT, 2024). Not all businesses require staff to log the time spent on tasks. It may be challenging for businesses to apportion part of a wage cost to fraud if the relevant staff member is working on multiple areas at once. While this survey followed best practice – the question wording explicitly asked how much staff would have got paid for this time and told respondents to include this cost even if dealing with these sorts of issues was already part of any staff member’s job role – it is possible the overall challenge of measuring staff costs led to this aspect being undervalued (see Home Office, 2025b).

Indirect costs
Approximately £59 million is estimated to be incurred in indirect costs by businesses as a consequence of fraud in the last 12 months. This includes:

• the value of lost share value, investors, assets or intellectual property
• the value expected from any lost customers or suppliers

Table 8: Costs as a consequence of fraud against businesses

Type of cost	Total cost (millions)	Unit cost
Financial loss	£507	£485
Lost output	£291	£278
Total consequence costs	£798	£763

Notes:

1. Figures may not sum due to rounding.

3.5 Costs in response to fraud

Response costs to fraud are incurred when action is taken following an incident of fraud. Total response costs of fraud against businesses are estimated to be £275 million, and this section summarises the costs.

3.5.1 Financial industry costs

Data from the financial industry estimates that costs in response of fraud total £770 million (YE March 2024 prices). This cost is apportioned to individuals and businesses in accordance with Action Fraud reporting splits by victim type. This analysis estimates that £78 million of this total is spent as a response of fraud against businesses.

Such costs include:

• resource costs for all post fraud touchpoints (for instance, operations teams, call centres, branch support)
• strategy costs, such as the implementation of risk-based payment delays, APP Reimbursement Requirement and the PSR’s Measure 1 data reporting
• industry engagement in response measures

3.5.1 Calculating CJS cost of fraud against businesses

See the individuals’ section for a breakdown of how total response costs have been calculated. PAS and Ministry of Justice data does not split by victim type. This analysis assumes that the total response cost as a result of fraud includes both fraud against individuals and businesses.

Total response costs have been apportioned to each cohort in accordance with the latest breakdown in Action Fraud data. Approximately 90% of Action Fraud reports are against individuals, with 10% reported to be against businesses. As a result, 90% of total response costs are estimated to be incurred through frauds against individuals and 10% for frauds against businesses.

The CJS is not the only avenue through which businesses that are victims of fraud pursue justice. Forms of civil justice such as private prosecutions and other out of court settlements are used by businesses to bring fraudsters to justice and attempt to recover some of the financial loss as a result of fraud incidents. Such costs have not been included in this analysis due to a lack of evidence of the magnitude of cases specific to fraud or the potential costs to firms pursuing justice through this avenue.

3.5.1 Aftermath costs

This analysis, using the midpoint method with ECS 2024 data, estimates that £148 million was incurred by businesses in medium term costs. These costs are incurred in response to an incident of fraud, including:

• any payments to third parties to investigate the fraud, run audits, or run training
• the cost of new or upgraded software or systems needed after the fraud
• recruitment costs if businesses had to hire someone new
• any legal fees, insurance excess, fines, compensation or PR costs related to the fraud

Table 9: Costs in response to fraud against businesses

Type of cost	Total cost (Millions)	Unit cost
Business aftermath costs	£148	£142
Finance industry response costs	£78	£75
Criminal justice system costs	£27	£26
Law enforcement costs	£21	£20
Total response cost	£275	£262

Notes:

1. Figures may not sum due to rounding

4. References

Curtis L and Burns A (2020) ‘Unit Costs of Health & Social Care 2020’, Personal Social Services Research Unit, University of Kent, Canterbury (viewed on 11 February 2026)

DBT (2024) ‘Business population estimates 2024’. Department for Business and Trade (viewed on 11 February 2026)

Dolan P, Loomes G, Peasgood T and Tsuchiya A (2005) ‘Estimating the intangible victim costs of violent crime’. British Journal of Criminology, volume 45(6), pages 958-976. Oxford University Press, Oxford (viewed on 12 February 2026)

DSIT (2024) ‘Cyber security breaches survey 2024’. Department for Science, Innovation and Technology (viewed on 12 February 2026)

Dubourg R, Hamed J and Thorns J (2005) ‘The economic and social costs of crime against individuals and households 2003/04’. Home Office Online Report 30/05 (viewed on 11 February 2026)

Heeks M, Reed S, Tafsiri M and Prince S (2018) ‘The economic and social costs of crime, second edition’. Home Office Research Report 99 (viewed on 11 February 2026)

HM Treasury (2022) ‘The Green Book’ (viewed on 11 February 2026)

Home Office (2023) ‘Fraud Strategy: stopping scams and protecting the public’. Policy Paper CP 839 (viewed on 11 February 2026)

Home Office (2024) ‘Economic Crime Survey 2024’ (viewed on 12 February 2026)

Home Office (2025a) ‘Police Activity Survey’ (viewed on 11 February 2026)

Home Office (2025b) ‘Economic Crime Survey 2024: Technical report’ (viewed on 12 February 2026)

MoJ (2024) ‘Criminal Justice System statistics quarterly: December 2023’. Ministry of Justice (viewed on 11 February 2026)

MoJ (2025) ‘Criminal court statistics quarterly: October to December 2024’. Ministry of Justice (viewed on 11 February 2026)

MoJ and HMPPS (2025) ‘Prison performance data 2023 to 2024’. Ministry of Justice and HM Prison and Probation Service (viewed on 11 February 2026)

NAO (2024) ‘The impact of fraud and error on public funds 2023-24’. National Audit Office (viewed on 11 February 2026)

NHS England (2021) ‘National Cost Collection for the NHS’ (viewed on 11 February 2026)

ONS (2020) ‘Nature of fraud and computer misuse in England and Wales: year ending March 2019’. Office for National Statistics (viewed on 11 February 2026)

ONS (2024) ‘Crime in England and Wales: Annual Trend and Demographic Tables’. Office for National Statistics (viewed on 11 February 2026)

Salomon JA, Haagsma JA, Davis A and others (2015) ‘Disability weights for the Global Burden of Disease 2013 study’ in The Lancet Global Health, volume 3(11), pages 712-723 (viewed on 11 February 2026)

Sentencing Council (2026) ‘Determinate prison sentences’ [online] (viewed on 11 February 2026)