Official Statistics

The cost of working age ill-health and disability that prevents work

Published 18 March 2025

Background

This analysis estimates the costs of working age ill-health or disabilities that prevent people working in the United Kingdom. It updates the data and methodology used in the 2016 Work, Health and Disability Green Paper Data Pack to look at the costs to the economy, government and individuals^{[footnote 1]}.

These estimates are not a definitive assessment of what savings might be achieved by further interventions to reduce ill-health or increase employment. The estimates in this costing can be regarded as the difference between current economic output, and the potential economic output if working age health conditions were not limiting or preventing work. This does not assume that all health issues could be resolved when entering work, rather that ill-health is no longer a barrier to work. The figures provided are not intended to provide a realistic view on the benefits that can be achieved by specific interventions but are illustrative of the cost to the economy and individuals of working age ill-health. Finally, the analyses do not estimate the impact of health on productivity at work.

The methods used to estimate the cost to the economy are each discussed in turn in the remainder of this section:

• lost production because of economic inactivity due to long-term or temporary sickness
• lost production due to sickness absence
• lost production due to informal care giving which removes people from the workforce
• additional costs to the NHS when someone’s health condition causes them to move from economically active to economically inactive
• lost Tax and forgone National Insurance returns to the Exchequer due to health conditions preventing or limiting employment
• cost of social security benefits related to health conditions that prevent people from working

This analysis is specific to 2022, due to data limitations and the requirement for consistency between the different categories^{[footnote 2]}.

This approach excludes costs of health conditions that are incurred regardless of whether a person is in-work or would continue to be incurred after they entered work. The following costs are also excluded:

• costs of working age inactivity due to working age people having informal caring responsibilities for people who are not of working age.
• costs of preventable working age mortality. Although future policy interventions may be able to mitigate these costs, available data does not allow a robust assessment of the cost of preventable mortality.

Method

A core part of the estimates for this work is the Gross Value Added (GVA) per person. GVA is measured as an average (mean) value per filled job and can be thought of as the individual’s job contribution to Gross Domestic Product (GDP). It encompasses more than the wage as an output, quantifying the wider value the individual brings to a business. It is reasonable to assume that the value of an individual’s output should exceed their wage, considering employment costs and a profit margin.

The GVA output per worker was £68,818 for the UK in 2022^{[footnote 3]}, ^{[footnote 4]}. This analysis also estimated a GVA figure for disabled people that may be more appropriate to apply to individuals with known ill-health. The GVA for a disabled worker is estimated for this analysis through multiplying mean GVA for all workers, by the ratio of median hourly pay of disabled workers^{[footnote 5]} to mean hourly pay for all workers (excluding overtime)^{[footnote 6]}, ^{[footnote 7]}. Using this, the estimated median annual GVA for a disabled person was £46,513 in 2022.

It is worth noting that GVA varies from region to region^{[footnote 8]}, and that ill-health that relates to work also has regional differences^{[footnote 9]}. However, this adds a complexity to the analysis that is beyond the scope of this analysis as regional splits are not available for the wide range of data required.

Lost production due to economic inactivity due to ill-health

This is the key element of the costings. The volumes of people reporting health-related inactivity are collected by the Labour Force Survey (LFS). In 2022 the number of people reporting inactivity due to long term sickness was 2.5 million, and 202,000 reported inactivity due to temporary sickness^{[footnote 10]}.

The central estimate here uses the mean GVA output per worker multiplied by the number of people who are economically inactive due to temporary sickness^{[footnote 11]} – it is assumed they will return to full productive capacity if they return to work – added to the number of economically inactive due to long term sickness multiplied with the estimated GVA for a disabled worker (assuming their condition persists within work). The central estimate is £132 billion.

To provide an upper estimate, the total number of individuals who were economically inactive was multiplied by the mean GVA output per worker, producing a total figure of £188 billion. A lower bound multiplied those who were economically inactive by the estimated disabled worker GVA, producing a figure of £127 billion.

The method used in the 2016 figure produced a range of between £73 to 103 billion using a comparable method. However, rather than using an estimated disabled worker GVA, the 2016 paper uses GVA adjusted by the ratio of median to mean earnings, while a higher estimate uses GVA per employee without that adjustment. The previous estimate also does not distinguish between temporary sickness and long term. Once inflation and the increase in the number of people economically inactive due to sickness have been accounted for, the 2016 figures are similar to those produced in this update.

Lost production due to sickness absence

ONS statistics record a total of 185.6 million days of sickness absence in 2022, around a third higher than in 2015^{[footnote 12]}. Assuming a 230-day working year based on a total of 6 weeks for statutory holidays and annual leave entitlement, this amounts to just over 800,000 full-time equivalent posts.

Using the average non-disabled worker GVA assumptions above, the estimated lost output is £56 billion. Not all this absence can be avoided: some will relate to conditions which cannot be prevented or remedied, and there will be a non-zero minimum for the level of sickness absence. However, there is no clear way to distinguish these absences, and even if deemed unavoidable, the costs entailed still result from working age ill-health that prevents work.

If we were to use the estimated disabled worker GVA for a lower bound, the cost is expected to be £38 billion.

The 2016 paper found the cost to be £15 to 20 billion. The method remains similar in this update, though for this update the same assumptions around sickness pay being indicative of further costs are not used, due to insufficient data in this area. The 2016 adjustment based of sickness pay reduced their figure from £20 to 30 billion to the £15 to 20 billion used. To further explain the difference in our outcome, sickness absence has increased by around a third, and GVA has increased by around £13,000 from 2016 to 2022^{[footnote 13]}.

Lost production due to informal caring

It is estimated that there were 5 million unpaid carers in the UK in 2022^{[footnote 14]}. Some of these carers will be working age, looking after someone who is working age, and out of work due to these unpaid caring responsibilities, and thus meet the definition of ‘cost of working age ill-health that prevents work’. In this instance it is the carer who is prevented from working, rather than the individual with the ill-health/health condition. The population of carers relevant to the economic cost of working age ill-health are: i) carers of working age who are out of work; ii) caring for people of working age; who iii) would work if they were not carers and whose caring responsibilities prevent them working. All these conditions must be met to be relevant for the present estimations.

The calculation is based on the number of working age unpaid carers that are currently unemployed, retired or otherwise economically inactive^{[footnote 15]}. The NHS Digital survey (2023 to 2024) can be used to determine the distribution of working age carers caring for someone of working age^{[footnote 16]}, and the distribution of out-of-work working age carers who are not in employment, because of their caring responsibilities, rather other reasons, such as retirement^{[footnote 17]}, ^{[footnote 18]}. Taking these distributions, around 550,000 carers were out of work due to working age ill-health (see Table 1).

Table 1: Assumptions used to estimate carers volume in 2022

Assumptions used to estimate carer volumes in 2022	Figures
Number of unpaid working age carers in the UK that are unemployed, retired or economically inactive	1,516,000
Of whom: caring for people of working age	55%
Of whom: would work if they were not carers and whose caring responsibilities prevent them working	65%
Estimated Carers in Scope	544,000

This figure is then multiplied by the mean GVA output per worker to produce a figure of £37 billion.

The 2016 paper produced a figure of £1 billion. The methodology used in this work has had slight refinement, so the difference is in part due to this. The 2016 paper took the percentage of the whole carers population who looked after someone who is working age, rather than the percentage of working age carers who looked after someone who is working age. This was also the case for the split of carers who are out of work due to their caring responsibilities. The key driver of the increase since the 2016 estimates is that the 2016 method assumed that only 5% of those being cared for would likely return to work, and that 95% of carers who fit the criteria would never be able to return to work as they would have to look after someone. The cost was therefore reduced by 95%. However, as there is still a cost to the care-giver and these estimates seek to consider all costs not just those which could be saved in practice, and as the informal care giver could enter employment if caring needs were addressed through social care, this analysis does not exclude any informal care givers that meet the criteria set out above.

Additional costs to the NHS

Limited sources exist to update the estimate of NHS costs included in the 2008 Dame Carol Black’s Review^{[footnote 19]} (£5 to 11 billion) and 2016 Green Paper published by the DWP. Much of the older methodologies are no longer replicable in 2022, for reasons including missing data and surveys that are no longer published. To manage the data constraints, the new method differs and looks at the additional treatment costs resulting from someone with a health condition moving from employment to inactivity, assuming that their health condition is preventing employment.

The OBR had published a cost in their Fiscal Report shown below that utilised the previous 2016 paper as a basis^{[footnote 20]}. The 2023 report states, “This results in additional costs [to the NHS] of £1,800 when people move from being active without health problems to health-related inactivity; £910 when people with health problems move from activity to inactivity; and £910 when economically inactive people develop work-limiting health conditions.”. The £910 figure was selected, as it isolates the cost from moving from employment with a health condition to economic inactivity with a health condition, assuming this is the cost of ill-health that prevents work, rather than just cost of ill-health. This is not true for the £1,800 figure that captures both the employment status change, but also health status change.

Table 2: Additional NHS costs per year when an individual with health problems moves from activity to inactivity

Additional NHS costs due to health-related inactivity	Cost
Additional costs per year when an individual with health problems moves from activity to inactivity	£910

The cost to the NHS of moving from economically active with a health condition, to economically inactive with a health condition is multiplied by the number of people who are inactive due to sickness, to estimate the extra cost to the NHS. This amounts to approximately £2 billion. The cost to social care was considered for inclusion, but if an individual’s social cares needs were the same in and out of work, if their health remained the same, the cost would be capturing the cost of ill-health, rather than ill-health that prevents work.

This figure is significantly below the 2016 figure (£7 billion) due to the methodological changes discussed above.

Lost flowbacks to exchequer

There are multiple ways to measure lost tax revenue. GVA accounts for tax gains, but it is complex to isolate the specific tax returns. Two methods have been utilised to isolate the tax returns, but these cannot be added to the GVA outputs (Sickness Absence, Economic Inactivity, and Informal Care Giving) as GVA accounts for tax.

The central estimate uses the same method as the 2016 paper. Lost output implies forgone tax revenues from individuals and employers. Tax revenues in scope are income tax, employer and employee national insurance (NIC), value added tax and corporation tax. A flowback rate is applied to lost output to estimate lost flowbacks. The flowback rate is obtained by estimating flowbacks from in-scope taxes and national insurance as a percentage of GDP using data published by the Commons Library^{[footnote 21]}. It states that the flowback rate averages 26.5% for 2023 to 2024. Applying this to the 2022 central estimates of lost output from health-related inactivity, sickness absence and informal care described earlier generates estimated lost flowbacks of around £57 billion.

An upper and lower bound of taxes excluding corporation tax was developed by using the Social Cost Benefit Analysis (SCBA) model maintained by the DWP, which can estimate the tax flowbacks for an individual working part time (18 hours) and full time (37 hours) at the 2022 median wage or median disabled wage^{[footnote 22]}. This model specifically includes Income Tax, Employees NIC, Employers NIC and Indirect Tax. It is assumed those who were economically inactive due to long-term sickness would earn the disabled person’s median wage, and those who were informal care givers, economically inactive due to temporary sickness, and those with sickness absence would earn the median wage ^{[footnote 23]}, ^{[footnote 24]}. Taking these cohorts, it was possible to multiply the flowbacks for the different wages and consolidating the flowbacks if the relevant individuals moved to part-time work as a lower bound of £14 billion and full-time work as an upper bound, £41 billion^{[footnote 25]}.

As a result, it assumed that the difference between the £57 billion from the GVA method and £14 to 41 billion, is missed corporation tax, which amounts to £16 to 43 billion, as corporation tax is the next remaining largest tax, and isn’t covered by the SCBA model (though other taxes may also form part of this figure).

The 2016 paper estimated this cost to be £21 to 29 billion. These updated estimates are considerably higher than this due to changes in method for the other metrics producing greater costs, increased tax rates, and greater numbers of inactive due to ill-health, sickness absence, and carers.

Table 3: The composition of exchequer flowbacks (Billions)

Exchequer flowbacks of which:	£ billion
Income Tax, NIC Employee, NIC Employer, Indirect Tax	14 to 41
Corporation Tax, and other tax	16 to 43
Total	57

Additional Benefit Payments

The social security system supports those who are unable to work due to ill-health, and those who care for those with ill-health. DWP publishes Benefit expenditure tables for benefits that support disabled people and people with health conditions from where welfare costs can be estimated^{[footnote 26]}.

For lower bound estimates, benefits were included that meet the definition of working age ill-health that prevents work. This includes incapacity benefits, carers benefits (reduced by the percentage of carers caring for someone of working age, assuming that those claim carers benefit are out of work due to their caring commitments), and housing benefits. Industrial Injuries Benefits were excluded as the number of working age claimants is relatively small. We are also unable to ascertain how many carers are claiming unemployment support due to their caring commitments, as some carers may also be out of work for other reasons, meaning we can only include the Carer Benefits.

For the upper bound estimates, the cost outlined above are added to the Personal Independence Payments, Armed Forces Independence Payment, and Disability Living Allowance costs for those who are also in receipt of incapacity benefits (i.e. out of work due to ill-health). This is around 66% of all PIP/DLA claimants^{[footnote 27]}. Whilst these do not meet the strict definition we have outlined; these benefit claims are highly likely to be resulting from the same ill-health that is preventing work.

This was taken from DWP published Benefit expenditure and Caseload for GB for Reserved Benefits, England & Wales Coverage for Devolved Benefits, specifically Table 4ii (Spring Budget). The benefits included are depicted in Table 4:

Table 4: Benefit expenditure resulting from working age ill-health that prevents work (Billions)

Benefit	£ billion
Armed Forces Independence Payment	Upper bound only	<1
Disability Living Allowance	Upper bound only	1
Personal Independence Payment (Adjusted to account for dual receipt of incapacity benefit)	Upper bound only	10
Employment and Support Allowance	Upper and lower bound	12
Incapacity Benefit	Upper and lower bound	<1
Income Support (incapacity / sick and disabled)	Upper and lower bound	<1
Universal Credit standard allowance and Health element	Upper and lower bound	10
Carer’s Allowance (adjusted for Working Age Carers, caring for Working Age people)	Upper and lower bound	2
Income Support for carers (adjusted for Working Age Carers, caring for Working Age people)	Upper and lower bound	<1
Universal Credit Carers Element (adjusted for Working Age Carers, caring for Working Age people)	Upper and lower bound	<1
Housing Benefit	Upper and lower bound	6
Universal Credit housing element	Upper and lower bound	4

Due to devolved benefit systems, there is not a unified table for UK expenditure. Select Scottish benefit lines have not been included, but reserved benefits are included in the wider figures e.g. UC and Housing Benefit^{[footnote 28]}. The upper, lower and midpoint were then scaled up by the relative increase in population from including Northern Ireland (assuming the same distribution of benefit expenditure across Northern Ireland^{[footnote 29]}). The midpoint is (£41 billion), taken from both the upper (£47 billion) and lower (£36 billion) bounds.

The 2016 paper reported a £19 billion cost associated with additional benefit payments. The updated figure is roughly double and can be explained by wider inflation increases and increased spend on incapacity benefits. The 2016 paper also reduces Carers benefits in their costings, assuming that many won’t return to work, but this updated analysis does not make this assumption as explained in the carers section. The 2016 report also halves housing support which this update does not do as the table used for this update is already limited to housing costs associated with Carers, disabled people, and those with health conditions^{[footnote 30]}.

Conclusion

The total cost of working age ill-health that prevents work in the UK is depicted in Table 5.

Table 5: Summary of the Cost of working age ill health and disability that prevents work (Billions)

Cost element	Description	Lower Bound £ billion	Upper Bound £ billion	Single/ Mid-Point £ billion
Economic inactivity	Lost output due to working-age ill-health which prevents work	127	188	132
Sickness absence	Lost output due to sickness absence	38	56	47
Informal care giving	Lost output due to working age carers caring for working age sick	37	37	37
NHS cost	The additional costs to the NHS when someone with a health problem is economically inactive rather than economically active	2	2	2
Benefits payments	Cost of social security benefits related to health conditions that prevent people from working	36	47	41
Total	Total expenditure in 2022	240	330	259
of which: Exchequer flowbacks	Tax and National Insurance foregone due to health-related worklessness	57	57	57

In total, the cost to the economy of working age ill-health and disability that prevents work in 2022 is estimated to be between £240-330 billion. The 2016 paper reported a cost range of £114-149 billion, meaning there has been an estimated increase of £126-181 billion in the cost of working age ill-health that prevents work in the six years since the original estimation. This is likely due to the methodological changes described above, but also increases in inflation, GVA, economic inactivity due to ill-health, sickness absence, and subsequent benefit support.

Statement of application of the Code of Practice for Statistics 

The analysis in this report has been produced as far as possible in line with the Code of Practice for Statistics. The code is built around 3 main concepts, or pillars, trustworthiness, quality and value: 

• trustworthiness – is about having confidence in the people and organisations that publish statistics 
• quality – is about using data and methods that produce assured statistics 
• value – is about publishing statistics that support society’s needs for information 

The following explains how we have applied the pillars of the Code in a proportionate way. 

Trustworthiness 

DWP analysts work to a professional competency framework and Civil Service core values of integrity, honesty, objectivity, and impartiality. Analysts have produced these statistics and conducted rigorous quality assurance in line to the standards usually applied to ad hoc releases. Background and methodology information is also included in the release. 

Care has been taken to ensure only those who needed to see the analysis prior to publication had access to it. 

The detailed methodology, data sources and approach taken in this research are set out in this report alongside the findings. The approach used builds on previous methodology used to calculate these estimations. 

Quality 

The process to produce the analysis in this report was conducted by professional analysts taking account of the latest administrative data and applying methods using their professional judgement. The analysis has been through a rigorous quality-assurance and sign-off process by other DWP analysts. 

Value 

The publication of this release ensures the information are equally available to all users as well as providing transparency. This research provides important new evidence for Ministers, policy makers and external stakeholders. 

1. Work, health and disability green paper: data pack – GOV.UK ↩

2. Some data for 2022 is missing, and this is flagged in this method document where appropriate. ↩

3. We have used GVA output for our GVA estimation, whereas the 2016 paper used balanced GVA (though for 2022 the balanced and output GVA were the same value). The output metric is based on output measures and not balanced against income or expenditure. The output measure is often used as the lead estimate for measurements such as GDP, as output measurement has a larger data content than income or expenditure approaches. See Recent challenges of balancing the 3 approaches of GDP – Office for National Statistics ↩

4. Output per worker, UK – Office for National Statistics – Table 12 (Accessed January 2025) ↩

5. Raw disability pay gaps, UK – Office for National Statistics – Table 1 (Accessed January 2025) ↩

6. Annual Survey of Hours and Earnings time series of selected estimates – Office for National Statistics – Table 1 (Accessed January 2025) ↩

7. Note – By using hourly wage we are not assuming any difference in hours worked by a disabled worker compared to the non-disabled worker. ↩

8. Regional gross value added (balanced) per head and income components – Office for National Statistics ↩

9. The employment of disabled people 2024 – GOV.UK ↩

10. A01: Summary of labour market statistics – Office for National Statistics – Table 11 (Accessed February 2025) ↩

11. This assumes temporary sickness lasts no less than one year as there is little to no clear evidence around temporary sickness length. ↩

12. Sickness absence in the UK labour market – Office for National Statistics – Table 2 (Accessed January 2025) ↩

13. Output per worker, UK – Office for National Statistics – Table 12 (Accessed January 2025) ↩

14. Stat Xplore FRS Individual Table 7 - Financial Year by Whether the Individual is an Informal Carer (Accessed March 2025) ↩

15. Stat Xplore FRS Individual Table 7 - Financial Year and Age band (16-64) of the Individual by Whether the Individual is an Informal Carer and Employment Status of the Adult (high level breakdowns) (Accessed March 2025) ↩

16. Personal Social Services Survey of Adult Carers in England, 2023-24 – NHS England Digital – Table T2a – Question 1 (Accessed November 2024) – Please note that we are unable to split the age of the person being cared for by employment status of the carer, so we assume that the distribution of unemployed working age carers caring for a working age person is the same as the distribution of working age carers caring for a working age person. ↩

17. Personal Social Services Survey of Adult Carers in England, 2023-24 – NHS England Digital – Table T2a – Question 21 (Accessed November 2024). We assume the age of the person being cared for has no impact on the carers ability to enter employment as question 21 cannot be split by age of person being cared for. ↩

18. The Personal Social Services Survey is England only, so we are assuming the same distribution across the rest of the UK based of England. ↩

19. Working for a healthier tomorrow: work and health in Britain – GOV.UK ↩

20. Fiscal risks and sustainability – July 2023 – Office for Budget Responsibility ↩

21. Tax statistics: an overview – House of Commons Library ↩

22. The DWP Social Cost-Benefit Analysis framework (WP86) – GOV.UK ↩

23. To note – We assume there are only tax implications for sickness absence when an individual receives SSP or less. This is estimated to be 38% of the population. Employee research Phase 2: Sickness absence and return to work. Quantitative and qualitative research. ↩

24. Raw disability pay gaps, UK – Office for National Statistics – Table 1 (Accessed January 2025) ↩

25. Calculations assume that workers receive paid holiday and full pay for a working year. These calculations are then converted into a monthly figure. ↩

26. Benefit expenditure and caseload tables 2024 – GOV.UK – Outturn and forecast tables: Spring Budget 2024 – table 4.(ii) (Accessed November 2024) ↩

27. Time series of incapacity and disability benefits overlaps -– February 2022 – England and Wales only, so it is assumed the UK has the same distribution. ↩

28. Excludes the following Scottish benefits - Adult disability payment, carers allowance supplement, and young carers grant. There is limited information on these lines, and they make up a comparatively small amount (£132 million). See - Social Security Scotland 2023 – 24 Budget (Accessed February 2025) ↩

29. Estimates of the population for the UK, England, Wales, Scotland, and Northern Ireland – Office for National Statistics – Mid-2022 edition – Table MYE1 (Accessed March 2025) ↩

30. They include spending on benefits explicitly directed at disabled people, people with health conditions and their carers; including Income Support and Housing Benefit paid to people in the disabled, incapacity and carer statistical groups, as well as the health related spend on Universal Credit. ↩