national resilience · the government's own risk list
The UK Chronic Risks Analysis: what the official statistics say about the Government's 26 chronic risks
In July 2025 the Cabinet Office and the Government Office for Science published the first Chronic Risks Analysis: 26 slow moving risks in seven themes, from fraud and antimicrobial resistance to skills shortages and climate change. It contains no scores and no time series. This page puts it against the numbers: which of the 26 the British Resilience Index measures, what the latest official statistics say where the document's evidence has aged, which of the index's most stressed indicators it never names, and which of its risks could be measured but are not.
How to read this
The Chronic Risks Analysis is a qualitative foresight document, not a dataset. Each risk gets a definition, a page of current evidence, example vulnerabilities and mitigations, and scenarios. Its figures are quoted here exactly as it prints them, with the page number and the source it cites; they are the Government's claims, not this site's. The index values are the September 2026 release of the British Resilience Index, data cutoff 25 August 2026. Where a figure the document quotes is set against a later official release, the later figure is transcribed from the publisher's own bulletin and linked, never computed. The coverage judgements (measured, partial, not measured) are this site's, and every one names the indicator it rests on so it can be checked.
26
chronic risks in the Government analysis
9
measured or partly measured by the index (5 direct, 4 partial)
17
not measured by the index
17
of its figures read again against the latest official release
Chronic Risks Analysis, Cabinet Office and Government Office for Science, published 8 July 2025 (132 pages, Open Government Licence v3.0). Index values from the British Resilience Index September 2026 release (data cutoff 25 August 2026). Comparison figures from ONS, Home Office, DWP, UKHSA, DfE, Skills for Care, the Environment Agency, Ofgem and the CMA releases dated October 2024 to August 2026, each linked where quoted. Built 11 September 2026.
the document
What the Chronic Risks Analysis is, and is not
The Chronic Risks Analysis is the Government's first assessment of slow moving risk, published on 8 July 2025 by the Cabinet Office and the Government Office for Science as the companion to the National Risk Register of January 2025, which covers acute emergencies. It identifies 26 chronic risks in seven themes and, for each, sets out a definition, current evidence, example vulnerabilities, example mitigations and what the future might hold. It was built by futures workshops and expert challenge rather than measurement, and it says so: it is not about predicting the future, and it is a snapshot in time that does not endeavour to present a regularly updated view. There are no scores, no rankings and no time series anywhere in its 132 pages.
8 July 2025
Published
26 in 7
Risks and themes
None
Scores, rankings or time series
Before the end of this Parliament
Next edition promised
The Resilience Action Plan's 2026 implementation report, published on 14 July 2026, mentions the analysis once and gives no date for the next edition. The Cabinet Office also runs a chronic risk dashboard, but it is a beta behind a departmental login and nothing from it is public. This page therefore treats the July 2025 document as the current one.
26 risks against 71 indicators
Which of the Government's chronic risks the index measures
The index scores 71 indicators in 11 domains, all from official series. Set against the 26 risks, 5 are measured directly, 4 are measured through a consequence or proxy the chapter itself names, and 17 are not measured at all. The split is not accidental. The index measures the condition of the country now, from the inside; most of the 26 are geopolitical, technological or biological risks that act on the country from outside, or that no official series yet counts. Each row shows the figure the document itself chose to highlight, then the indicators the judgement rests on with their September 2026 value and stress band.
Measured 5Partial 4Not measured 17
Security
Risk
The document's own headline figure
Index coverage
Changes in the nature of terrorism
page 15
Nine terrorist attacks in the UK between 2018 and 2023, killing six people and injuring 20
CONTEST 2023, cited on page 16
Not measured
No index indicator. The Home Office publishes terrorism arrests quarterly; see the gaps section.
Changes in the nature of serious and organised crime
Adult social care vacancies: 6.2% of adult social care roles vacant, 2025-26 · Stable (19)
The index scores the labour market outcomes the chapter worries about, and the social care vacancy rate it quotes, but has no measure of skills mismatch itself.
Coverage judgements are this site's. The indicator values and bands are read from the September 2026 release at build time. A CRA figure marked as commercial or a briefing is quoted because the document chose it, not because this site treats it as an official statistic.
the same measures, a year on
Where the numbers have moved since the document was written
The analysis is dated July 2025 but its evidence is older: much of it is from 2022 to 2024, and some of it was superseded before publication. These 17 cards take a figure the document prints and set the latest official release of the same measure beside it. 5 have risen, 5 have fallen, 1 is unchanged, 4 had already been overtaken by a newer official figure when the document went to print, and 2 sit on a series whose basis has changed. Every later figure is transcribed from the publisher's bulletin and linked.
Fraud and illicit finance
Risen since
Fraud incidents, Crime Survey for England and Wales
The document says
3.9 million fraud offences, year ending June 2024ONS, cited on page 23
ONS reports no statistically significant change on the 4.2 million of the year before. The series has gone 3.2 million (year ending March 2024), 4.2 million, 4.5 million, so the CRA caught it on the way up.
Serious and organised crime
Fallen since
Police recorded firearms offences, England and Wales
The document says
6,268 offences, up 6 per cent, year ending March 2024NCA National Strategic Assessment 2025, cited on page 19
The 2023 figure of 5,448 had already broken the 2022 record when it was published in October 2024, nine months before the CRA. The record has now been set three years running.
Serious and organised crime
Risen since
Potential victims of modern slavery referred to the National Referral Mechanism
The document says
4,758 referrals in a quarter, July to September 2024Home Office, cited on page 20
The latest quarter is 13 per cent above the same quarter a year earlier. The Home Office warns that some modern slavery publications are delayed by an IT system change.
Changes in the nature of cyber security threats
Unchanged
Businesses identifying a cyber breach or attack in the last 12 months
The document says
43 per cent of UK businesses, Cyber Security Breaches Survey 2025DSIT, cited on page 29
Latest official figure
43 per cent of businesses; NCSC nationally significant incidents 204, from 89 the year before, Cyber Security Breaches Survey 2025/2026 (published April 2026) and NCSC Annual Review 2025DSIT and Home Office; NCSC ↗
The share of businesses attacked is flat. The count of incidents the NCSC judges nationally significant more than doubled. Both are reconciled on this site's cyber security page.
Climate change
Superseded before publication
Properties in areas at risk of flooding, England
The document says
5.5 million homes and businesses, 2023Environment Agency Flood Action Week release, cited on page 55
Latest official figure
About 6.3 million properties, National assessment 2024, first published 17 December 2024Environment Agency, NaFRA2 ↗
The Environment Agency's new national assessment was published seven months before the CRA and is the figure this site's flood risk page uses. The CRA cites the older press release.
Impacts of ongoing skills shortages and mismatches
Fallen since
Vacant posts in adult social care, England
The document says
122,000 vacancies, a vacancy rate of 8 per cent, House of Commons Library briefing, 2020Cited on page 123
The CRA's headline for the whole skills chapter rests on a five year old briefing. Vacancies had peaked and fallen twice over by the time it was published. The index tracks Skills for Care's monthly series, at 6.2 per cent.
Impacts of ongoing skills shortages and mismatches
Risen since
UK unemployment rate, people aged 16 and over
The document says
4.4 per cent, December 2024 to February 2025ONS, cited on page 123
A chapter about shortages of workers now sits beside a labour market with fewer vacancies and more unemployment than when it was written.
Impacts of ongoing skills shortages and mismatches
Fallen since
Employers providing training, and skill shortage vacancies
The document says
60 per cent of employers trained staff in the last 12 months, against 65 to 66 per cent in 2011 to 2017, Employer Skills Survey 2022DfE, cited on page 123
Latest official figure
59 per cent trained staff; 250,500 skill shortage vacancies, less than half the 531,200 of 2022; 6 per cent of employers had one, down from 10 per cent, Employer Skills Survey 2024, published 24 July 2025DfE ↗
Published 16 days after the CRA. Training kept slipping, but skill shortage vacancies halved, the first fall in the series.
Impact of demographic change
Superseded before publication
Total fertility rate, England and Wales
The document says
1.58 children per woman in 2020, the lowest since records began in 1938; 1.61 in 2021, 2020 and 2021ONS Births in England and Wales 2021, cited on page 73
Latest official figure
1.44 in 2023, the lowest since records began; 594,677 births in 2024, with the 2024 rate not yet calculated, 2023 rate (published 28 October 2024) and 2024 births (published 1 July 2025)ONS, Births in England and Wales ↗
The CRA's record low had been beaten twice, in 2022 (1.49) and 2023 (1.44), before it was published. The fall it describes is steeper than it says.
Impact of demographic change
Risen since
UK population
The document says
67 million, up 3.7 million (5.9 per cent) on mid 2011, mid 2021Cited on page 73
Three years of growth after the CRA's baseline added about 2.3 million people, more than half the whole increase of the previous decade.
Impact of demographic change
Series revised
Net migration, UK
The document says
An annual average of 200,000, 2011 to 2021Cited on page 74
Latest official figure
171,000, down from 331,000 the year before and a peak of 944,000 in the year ending March 2023, year ending December 2025, provisionalONS, Long-term international migration ↗
A decade average that stops in 2021 misses the peak that followed and the fall after it. The chapter was written after both had been published.
Reliance on global supply chains
Fallen since
Ofgem price cap, typical dual fuel household
The document says
The cap rose from 1,216 pounds in winter 2021/22 to 4,059 pounds in winter 2022/23; the Energy Price Guarantee limited bills to 2,500 pounds, 2022/23ONS, cited on page 115
Latest official figure
1,663 pounds a year (1,862 pounds on the earlier usage basis), while household energy debt and arrears reached a record 4.79 billion pounds, July to September 2026 cap; debt at first quarter 2026Ofgem price cap levels and debt and arrears indicators ↗
The shock the CRA uses as its worked example has passed through the bill and settled into debt. Both series are on this site's energy bills page.
Antimicrobial resistance
Risen since
Antibiotic resistant infections, England
The document says
Severe antibiotic resistant infections up 6.6 per cent between 2020 and 2022, to nearly 58,000, 2022UKHSA ESPAUR 2022 to 2023, cited on page 88
The two measures differ: the CRA's counts all severe resistant infections, the latest headline counts bloodstream infections only, so the levels are not comparable. The direction is the same: still rising, with use falling.
Dominance of global technology companies and concentration of risk
Superseded before publication
Competition in UK cloud computing
The document says
Two companies control 70 to 90 per cent of the UK cloud market, 2023, provisional findingsCMA cloud services market investigation, cited on page 37
Latest official figure
The CMA's final decision recommended strategic market status investigations into Microsoft and AWS; in March 2026 it accepted their commitments on egress fees and interoperability and opened an investigation into Microsoft's business software instead, 31 July 2025 and 31 March 2026CMA ↗
The CRA quotes the provisional findings. The final decision came three weeks after it, and the regulatory outcome eight months after that.
Disproportionate impact on vulnerable persons
Series revised
People in relative low income after housing costs, UK
The document says
14.3 million people, 21 per cent, a decrease of 0.1 million, 2022/23DWP Households Below Average Income, cited on page 78
DWP says there is a break in the series, so the CRA's 21 per cent and the new 20 per cent are not on the same basis. On the old basis the rate was unchanged at 21 per cent a year after the CRA's figure.
the index's worst indicators
The stresses the document does not name
The September 2026 release places 9 indicators in the high or extreme stress bands. Read against the text of the analysis, 2 are named, 2 appear only in passing, and 5 do not appear at all. That is a difference of purpose more than a disagreement: the document assesses risks that could erode the country over decades, the index measures pressures that are visible in the official statistics now. But suicide, homelessness and the court backlog are chronic by any reading, and the Government's own chronic risk document has no place for them.
Indicator
Latest value
Stress
In the document
Suicide rate
Population Health
11.4 age-standardised deaths per 100,000
2024 · deteriorating
97 Extreme stress
Not named
The word suicide does not appear in the document.
Common mental-health condition prevalence
Population Health
22.6% of adults aged 16-64
2023/4 · deteriorating
87 High stress
In passing
Mental health appears six times, as a consequence of social media use, terrorism, animal disease or the cost of living, never as a measured condition of the population.
Rough sleeping snapshot (England)
Housing and Household Formation
4,793 people
Autumn 2025 · deteriorating
85 High stress
Not named
Neither rough sleeping nor homelessness appears.
Crown Court outstanding caseload (E&W)
Public Service Capacity
80.1 thousand open cases
End Mar 2026 · deteriorating
84 High stress
Not named
The only court in the document is the International Criminal Court; the word backlog appears once, about care.
Households in temporary accommodation (England)
Housing and Household Formation
135.6 thousand households
Q1 2026 · deteriorating
83 High stress
Not named
Temporary accommodation does not appear; housing appears three times, as a policy area.
Drug-poisoning deaths
Population Health
93.9 deaths per million population (age-standardised)
2024 · stable
83 High stress
One sentence
Named once, in the serious and organised crime chapter: drug related deaths at the highest level since records began in 1993.
Fraud incidents (CSEW)
Rule of Law and Safety
4.5 million incidents per year
YE Mar 2026 · deteriorating
82 High stress
Own chapter
Fraud and illicit finance is one of the 26 risks, with its own chapter.
EHC plan timeliness
Childhood and Social Floor
46.1% of new EHC plans issued within 20 weeks
2025 · stable
81 High stress
Not named
Special educational needs and education, health and care plans do not appear.
Inactivity due to long-term sickness
Work and Productivity
2.792 million people
2025 · stable
78 High stress
In passing
Inactivity appears once, as the trend the Get Britain Working white paper aims to reverse; sickness twice, as staff sickness leave.
Word counts are from the extracted text of the PDF, so terms that appear only inside a diagram would be missed; the impact maps were read by eye for these nine. Stress bands: High stress is a score of 75 to 89, Extreme stress 90 to 100.
the cross cutting analysis
What the document says is most connected, and how much of it the index can see
The CRA closes with an impact map of how its 26 risks act on one another and on the acute risks in the National Risk Register. It names the most connected risks in prose rather than with a table, so the figures below are the ones it prints.
11
direct impacts on acute risks from state threats, the largest total of direct connections
19
risks and vulnerabilities affected by the dominance of global technology companies, itself affected by only two
14
risks and vulnerabilities affected by competition for critical minerals, itself affected by only two
The five risks the document calls most connected
Challenges to international institutions: the largest number of direct impacts on other risksNot measured
State threats: the largest total of direct connections, incoming and outgoingNot measured
Changes in the nature of terrorism: tied second largest total of connectionsNot measured
Changes in the nature of cyber security threats: tied second largest total of connectionsNot measured
Climate change: tied second largest total of connectionsMeasured by the index
Of the five risks the document calls most connected, the index measures one, climate change, through territorial emissions. The document also names terrorism, serious and organised crime and vulnerable persons as the risks most acted on by the others; the index measures the third of those most heavily of all 26, with eight indicators.
what is not measured, and whether it could be
The 17 risks outside the index, and the official series that exist for them
For 3 of the 17 unmeasured risks a regular official series already exists, and one of them, terrorism, shows why an indicator of chronic risk has to be chosen with care. For 6 more there is a survey, a register or a case list that could support an indicator. For 8 there is nothing regular to count. Figures are quoted only where this site has read them from the publisher.
Official series 3Survey or register 6No series 8
Security
Changes in the nature of terrorism
Official series exists
Home Office, Operation of police powers under the Terrorism Act 2000, published quarterly.
3,061 arrests for terrorism related activity in the year ending March 2026, up from 238 the year before; 2,819 of them, 92 per cent, linked to supporting Palestine Action after its proscription on 5 July 2025. 273 people in custody for terrorism connected offences at 31 March 2026.
DSIT and Home Office Cyber Security Breaches Survey, annual; NCSC Annual Review incident counts.
43 per cent of businesses identified a breach or attack in the 2025/2026 survey; the NCSC handled 204 nationally significant incidents in the year to August 2025.
The Animal and Plant Health Agency and Defra publish confirmed cases of notifiable diseases such as avian influenza as they occur.
Biosecurity
Foodborne diseases
A survey or register exists
UKHSA publishes laboratory reports of the main gastrointestinal pathogens; the Food Standards Agency's burden estimate the CRA cites is for 2018.
Biosecurity
Plant pests
A survey or register exists
Defra's UK Plant Health Risk Register lists and scores pests, but is a register rather than a count over time.
Biosecurity
Impacts from the expansion of engineering biology
No official series
No official series.
Biosecurity
Increasing collection and use of human biological data
No official series
No official series.
Economic
Reliance on global supply chains
No official series
No regular series; the ONS Business Insights and Conditions Survey asked about supply disruption during 2022 and 2023.
Economic
Impacts from emerging financial systems
A survey or register exists
The Financial Conduct Authority runs periodic cryptoasset consumer research.
questions this raises
Questions this raises
1.
The foreword promises a new Chronic Risks Analysis before the end of this Parliament. Should the next edition carry a measure, even a single indicator, for each of the 26 risks, so that the edition after it can say which got worse?
2.
Seventeen of the 26 risks have no place in a domestic resilience index. Which of them could be measured from series that already exist: terrorism arrests, resistant bloodstream infections, cyber breaches, modern slavery referrals?
3.
Nine indicators sit at high or extreme stress in the September 2026 release and only two of them are named in the document. Is that because they are pressures on services rather than chronic risks, or because the document looks outward while the measured stress is at home?
4.
Several of the document's figures had already been superseded when it was published: the 5.5 million flood figure, the 1.58 fertility rate, the 2020 social care vacancy count, the 2022 drug deaths record. What does a snapshot in time mean when the evidence behind it is two to five years old?
5.
Terrorism arrests rose from 238 to 3,061 in a year because of one proscription. When a series can move by 1,186 per cent on a legal decision, what should an indicator of chronic risk count?
6.
The Cabinet Office runs a chronic risk dashboard behind a departmental login. Should it be public?
The Chronic Risks Analysis is the Government's document. Every figure attributed to it is quoted as it prints it, with the page number and the source it cites. Those sources range from ONS statistics to commercial estimates and a 2020 briefing; the crosswalk labels the ones that are not official statistics.
Coverage is a judgement made by this site: measured means at least one scored indicator tracks the risk itself; partial means the index scores a consequence or a proxy the chapter itself names; not measured means neither. Indicators the index publishes as provisional placeholders are not counted.
The index measures domestic conditions on official series. Most of the 26 risks are geopolitical, technological or biological and are outside its scope by design, so the gap count is a statement of scope, not a finding that the risks are unimportant.
Where a later official figure is set beside a CRA figure the two are not always the same measure. The antimicrobial resistance headline changed from all severe resistant infections to bloodstream infections, and DWP rebuilt the low income series with administrative data. Those cards say so and draw no comparison of levels.
The word counts in the unnamed section come from the text of the PDF extracted with pdftotext. Diagrams rendered as images are not searchable, so a term shown only in an impact map would be missed. The impact maps were read by eye for the nine indicators listed.
The Cabinet Office chronic risk dashboard is a beta behind a departmental login. Nothing from it appears here, because nothing from it is public.
Index values are those of the September 2026 release and carry the vintage of each indicator's official publication. The crosswalk shows the period beside each figure.
Every figure on this page traces to its official source. Share it as it stands.
One email a month: the national resilience score, what moved, and the newest analyses like this one, plus The Quarterly Record four times a year. Every figure traced to its official source.