The Global Risks Index · 46 entries · 2026-10
The risk index landscape
What the world’s risk indices measure, how, and where they stop, so NXS-GRX (canonical data, provenance and risk) can interoperate with them and close the gaps between them. Source series are read where licences admit and never re-weighted.
页面 GRX-LND版本 2026 基线版证据截止 2026-09-30数值 尚未发布方法 v1.0.0-draft
FIG 1 · Families read
9 entries
多危害灾害与人道主义风险
INFORM Risk · INFORM Severity, Climate Change, Warning and subnational models · WorldRiskIndex.
Open
10 entries冲突、和平、脆弱与治理
Fragile States Index · Global Peace Index and Global Terrorism Index · ACLED Conflict Index.
Open
7 entries经济、主权和金融风险
Country Risk Classification · Debt sustainability frameworks (LIC-DSF, SRDSF) · Sovereign credit ratings.
Open
4 entries气候、环境与自然
Environmental Performance Index · Central-bank network climate scenarios · Aqueduct water risk.
Open
4 entries卫生与粮食
Global Health Security Index · IHR self-assessment (SPAR) and Joint External Evaluation · Integrated Food Security Phase Classification and Cadre Harmonisé.
Open
4 entries技术、网络与前沿风险
Global Cybersecurity Index · National Cyber Security Index · AI indices and incident monitors.
Open
3 entries感知与前瞻
Global Risks Report · World Risk Poll · Corporate risk barometers and reinsurer research.
Open
5 entries发展与韧性
Human Development Index and Multidimensional Poverty Index · SDG Index and Social Progress Index · Sendai Framework Monitor.
打开
46
indices read
8
families
17
gaps addressed
9
standards followed
01Gaps
What no index does yet
Established risk and resilience indices, read for interoperability and validity checks. Names are those of their stewards and are cited for reference only; no text of any index is reproduced and no endorsement is implied. Commercial providers are described by category. Indicator counts change between editions; the current methodology note of each index governs. NXS-GRX (canonical data, provenance and risk) relation: SOURCE (a published series NXS-GRX may read as a source where its licence admits, never re-weighted), CROSSWALK (used for validity checks and mapping), COMPLEMENT (a different object NXS-GRX does not reproduce).
| 差距 | NXS-GRX design |
|---|---|
| No index unifies hazard, vulnerability, conflict, macro-finance, cyber, health, food, nature and technology in one structure | Twenty index domain indices in four pillars, one grammar |
| Systemic and cascading risk is described, not modelled | Coupling instrument: typed dependency networks, contagion models, coupled beside standalone composite |
| 次国家层面覆盖薄弱且不可比 | ADM1, ADM2 and one H3 fabric for hazard, exposure and capacity |
| 很少报告不确定性；排名被过度解读 | 90% intervals, class probabilities and Sobol sensitivity on every value; no league tables |
| 数据滞后一至三年 | Pulse instrument: weekly nowcast over admitted event streams |
| 很少有指数进行预测或回测 | 展望工具，附公开的验证评分卡 |
| 情景预测稀少且仅限宏观层面 | Pathways instrument: SSP × warming level and transition families to 2030, 2050 and 2100 |
| 感知风险与测量风险混为一谈 | Perception instrument kept separate, with a published gap |
| Values rarely trace to sources, versions and licences | 记录级来源追溯、许可关口、公开更正日志 |
| 机器访问能力弱（PDF 和电子表格发布） | API, SDMX keys, vector tiles, bulk files and versioned releases |
| 许可限制阻碍衍生再利用 | 按来源的许可关口；扣留的值计入统计 |
| Coverage gaps cluster in fragile and small-island states | Coverage vector per country; imputation flagged and capped |
| 纸面上的能力被误认为实际表现 | Capability ladder from listed to demonstrated in real events |
| 方法变更破坏可比性 | 语义化版本、回溯序列、每次发布的摘要 |
| Neglected domains: AI, space weather, biosecurity, minerals, digital infrastructure, tail risk | Exponential technology and strategic systems series; Tail instrument |
| 各指数之间没有共享分类体系 | hazards with HIPs crosswalks; indicators with 501 crosswalks to source code lists |
| 平均值掩盖了厚尾 | Exceedance curves, return-period losses and expected shortfall where loss can be modelled |
02类别
多危害灾害与人道主义风险
| 指数 | Steward | 覆盖范围 | 周期 | Structure and method | Relation | Limitation |
|---|---|---|---|---|---|---|
| INFORM Risk | EU Joint Research Centre with the IASC reference group | 191 countries | Annual with mid-year update | 3 dimensions, about 15 components, roughly 74 to 80 indicators. Arithmetic means to component level, geometric mean of hazard and exposure, vulnerability and lack of coping capacity; 0–10 | Source series | No financial, cyber or systemic risk; no intervals; one to three year data lag |
| INFORM Severity, Climate Change, Warning and subnational models | EU Joint Research Centre and partners | Crises, projections, admin levels | Monthly to irregular | INFORM grammar applied to severity, projections and subnational units. INFORM method | 对照 | Subnational models are not mutually comparable and share no grid |
| WorldRiskIndex | NGO alliance with a university institute | 193 countries | 年度 | Exposure × vulnerability (susceptibility, coping, adaptation); more than 100 indicators. Geometric mean; 0–100 | 对照 | Population-driven exposure; no conflict hazard |
| Global Assessment Report and Global Risk Assessment Framework | UNDRR | 全球 | About biennial | Thematic assessment; earlier probabilistic global risk model. Probabilistic loss modelling and narrative | Complement | No standing index or API; systemic risk described rather than measured |
| ThinkHazard! | World Bank GFDRR | Admin-2, global | Occasional | 11 hazards by intensity and frequency. Classification in four classes | 对照 | Hazard only; no vulnerability |
| Climate Risk Index | Germanwatch (NGO) | About 170 to 190 countries | 年度 | 6 loss indicators. Weighted rank sum; single-year and 30-year windows | 对照 | Backward-looking; inherits loss-database reporting bias |
| ND-GAIN | University of Notre Dame | About 185 countries | 年度 | Vulnerability and readiness; 45 indicators. Arithmetic means; 0–100 | 对照 | Country level only; readiness bias toward high-income countries |
| States of Fragility | OECD | 170+ contexts screened | About biennial | 6 dimensions, about 50 indicators. Principal components and clustering into classes | 对照 | Classifies rather than measures; not predictive |
| Global Conflict Risk Index | EU Joint Research Centre | About 130 to 190 countries | Annual or semi-annual | About 24 structural variables. Logit models of conflict probability; 0–10 | 对照 | Structural only; no event nowcast |
03类别
冲突、和平、脆弱与治理
| 指数 | Steward | 覆盖范围 | 周期 | Structure and method | Relation | Limitation |
|---|---|---|---|---|---|---|
| Fragile States Index | Fund for Peace (NGO) | 179 countries | 年度 | 12 indicators in 4 groups. Content analysis, quantitative data and review; 0–120 | 对照 | Media-volume bias; reuse restricted |
| Global Peace Index and Global Terrorism Index | Institute for Economics and Peace (NGO) | About 163 countries | 年度 | 23 indicators; 4 terrorism indicators. Banded weighted means; log-banded scores | 对照 | Contested weights; source-dataset dependency |
| ACLED Conflict Index | ACLED (NGO) | About 240 countries and territories | Annual and weekly | Deadliness, danger, diffusion, fragmentation. Rank-based severity categories | Source series | Licence limits redistribution |
| UCDP | Uppsala University | Global, geocoded events | Annual and monthly | Event data with best, low and high fatality estimates. Event coding; no composite | Source series | Threshold of 25 battle deaths for conflicts |
| V-Dem | University of Gothenburg | 202 countries | 年度 | More than 500 indicators, 5 high-level indices. Bayesian measurement model with credible intervals; 0–1 | 对照 | Expert subjectivity; one-year lag |
| Worldwide Governance Indicators | World Bank | About 214 economies | 年度 | 6 dimensions from about 35 sources. Unobserved components with standard errors | Source series | Perception-based; correlated source errors |
| Corruption Perceptions Index | Transparency International (NGO) | 180 countries | 年度 | 13 expert and business sources. Standardised mean with confidence intervals; 0–100 | 对照 | No-derivatives licence constrains reuse |
| Global Organised Crime Index | GI-TOC (NGO) | 193 countries | Biennial | Criminal markets, actors and resilience. Expert assessment; 1–10 | 对照 | Expert-driven |
| Basel AML Index | Basel Institute on Governance (NGO) | About 160 to 200 jurisdictions | 年度 | 17 to 18 indicators in 5 domains. Weighted mean; 0–10 | 对照 | Depends on evaluation cycles |
| Bertelsmann Transformation Index and Ibrahim Index of African Governance | 基础 | 137 and 54 countries | Biennial | Expert criteria; about 96 indicators. Expert scores; arithmetic means | 对照 | Partial coverage |
04类别
经济、主权和金融风险
| 指数 | Steward | 覆盖范围 | 周期 | Structure and method | Relation | Limitation |
|---|---|---|---|---|---|---|
| Country Risk Classification | OECD (export credits arrangement) | About 200 countries | Three times a year | Model plus expert adjustment. Classes 0–7 | 对照 | Transfer and convertibility risk only; model unpublished |
| Debt sustainability frameworks (LIC-DSF, SRDSF) | IMF and World Bank | Low-income and market-access countries | Per review | Thresholds, stress tests, fan charts. Mechanical signals plus judgement | Source series | Per-country, not a comparable index; NXS-GRX never reproduces the published signals |
| Sovereign credit ratings | Commercial rating agencies (category) | About 130 to 150 sovereigns | Ongoing | Committee scorecards. Letter grades | Complement | Paywalled; NXS-GRX is not a rating and uses no rating scale |
| Early-warning indicators and financial stress indices | BIS and central banks | 40+ economies | Daily to quarterly | Credit gaps, debt service, stress composites. Gaps, thresholds, portfolio weights | Source series | Advanced economies mostly |
| Global Supply Chain Pressure Index | Federal Reserve Bank of New York | 全球 | Monthly | 27 variables. Principal components; standard deviations | 对照 | No country decomposition |
| Logistics Performance Index | World Bank | 139 countries | Every 3 to 5 years | 6 dimensions plus tracking KPIs. Survey; 1–5 | 对照 | Infrequent; perception-based |
| Economic complexity | University research centres | About 130 to 145 countries | 年度 | Export diversity and ubiquity. Eigenvector method | 对照 | Structural signal, not risk |
05类别
气候、环境与自然
| 指数 | Steward | 覆盖范围 | 周期 | Structure and method | Relation | Limitation |
|---|---|---|---|---|---|---|
| Environmental Performance Index | Yale and Columbia universities | 180 countries | About biennial | 58 indicators, 11 issue categories. Weighted mean; 0–100 | 对照 | Performance, not risk |
| Central-bank network climate scenarios | Network of central banks and supervisors | Global, regional, country | About annual | 6 to 7 scenario families. Integrated assessment and macro models | Source series | Pathways, not an index; physical risk understated |
| Aqueduct water risk | World Resources Institute (NGO) | Global sub-basins | Occasional | 13 indicators. Weighted; 0–5 with projections | Source series | Coarse units; ageing layers |
| Biodiversity intactness, ENCORE and the Ocean Health Index | Museums, UNEP-WCMC, universities | Global grid, sectors, EEZs | Varies | Models, dependency matrices, goals. Statistical models; materiality ratings; 0–100 | 对照 | Weak link to local ground truth |
06类别
卫生与粮食
| 指数 | Steward | 覆盖范围 | 周期 | Structure and method | Relation | Limitation |
|---|---|---|---|---|---|---|
| Global Health Security Index | NGO and university partnership | 195 个国家 | 2019 and 2021 | 6 categories, 37 indicators. Weighted mean; 0–100 | 对照 | Measured capacity on paper; not updated since 2021 |
| IHR self-assessment (SPAR) and Joint External Evaluation | WHO | 196 States Parties | Annual and ad hoc | 15 capacities; 19 technical areas. Self and external scoring; 1–5 | Source series | Self-reporting bias |
| Integrated Food Security Phase Classification and Cadre Harmonisé | Multi-agency technical partnership | Assessed areas | Per country cycle | Phases 1–5, current and projected. Evidence convergence by technical working groups | Source series | Limited to assessed areas |
| Global Hunger Index and Global Report on Food Crises | NGOs; FAO, WFP and EU network | About 50 to 136 countries | 年度 | 4 indicators; headcounts. Standardised sum; aggregation | 对照 | Three-year lag; aggregation rather than scoring |
07类别
技术、网络与前沿风险
| 指数 | Steward | 覆盖范围 | 周期 | Structure and method | Relation | Limitation |
|---|---|---|---|---|---|---|
| Global Cybersecurity Index | ITU | 194 countries | About biennial | 5 pillars. Questionnaire; 5 tiers since 2024 | Source series | Measures commitment, not threat or incidents |
| National Cyber Security Index | Government-backed academy | 170+ countries | Rolling | 12 capacities, 46 indicators. Evidence-checked; 0–100 | 对照 | Capacity only |
| AI indices and incident monitors | Universities, OECD and NGOs | Varies | Annual or continuous | Descriptive indicators, responsible-AI scores, incident data. Mixed | 对照 | No probabilistic national AI-risk index exists |
| Global catastrophic risk assessments | University centres and foundations | 全球 | Irregular | Narrative and elicited probabilities. Expert elicitation | Complement | No standardised country index |
08类别
感知与前瞻
| 指数 | Steward | 覆盖范围 | 周期 | Structure and method | Relation | Limitation |
|---|---|---|---|---|---|---|
| Global Risks Report | World Economic Forum (not-for-profit foundation) | 全球 | 年度 | Survey of experts rating about 33 global risks over 2- and 10-year horizons. Perception survey | 对照 | Perception only; no probabilities; respondent pool skewed |
| World Risk Poll | Lloyd’s Register Foundation (charity) | About 142 countries | Biennial | Experienced and perceived risk. Household poll | 对照 | Survey cadence |
| Corporate risk barometers and reinsurer research | Commercial insurers and reinsurers (category) | 全球 | 年度 | Surveys and loss databases. Perception and proprietary loss data | Complement | Proprietary inputs |
09类别
发展与韧性
| 指数 | Steward | 覆盖范围 | 周期 | Structure and method | Relation | Limitation |
|---|---|---|---|---|---|---|
| Human Development Index and Multidimensional Poverty Index | UNDP with OPHI | 193 and about 110 countries | 年度 | 3 dimensions; 10 indicators. Geometric mean; Alkire–Foster | Source series | Survey lags |
| SDG Index and Social Progress Index | UN SDSN; a non-profit | About 167 to 170 countries | 年度 | 17 goals; 12 components. Arithmetic means; PCA weights | 对照 | Outcome dashboards, not risk |
| Sendai Framework Monitor | UNDRR | Reporting countries | 年度 | 38 indicators under 7 targets. National self-reporting | Source series | Uneven reporting |
| Global Innovation Index | WIPO | About 133 countries | 年度 | 7 pillars, about 80 indicators. Arithmetic means with a published statistical audit | 对照 | Innovation, not risk; a model of audit transparency |
| Business Ready (B-READY) | World Bank | 101 economies (2025), expanding | 年度 | 3 pillars over 10 topics. Expert questionnaires and firm surveys; 0–100 | Source series | Coverage still expanding |
10标准
Standards NXS-GRX follows
STDOECD/JRC Handbook on Constructing Composite IndicatorsThe ten construction steps NXS-GRX follows→ STDJRC COIN statistical audit practiceCorrelation structure, coherence, Monte Carlo uncertainty and rank stability→ STDISO 31000 and IEC 31010Risk-management vocabulary and assessment techniques (referenced; no conformance claimed)→ STDSendai Framework for Disaster Risk ReductionTargets A–G and 38 indicators as crosswalk keys→ STDUNDRR–ISC Hazard Information Profiles (2025)Canonical hazard keys through Hazard Standard crosswalks→ STDIRDR peril classificationCrosswalk for loss databases→ STDSDMX (ISO 17369)Observation keys, code lists and statistical exchange→ STDUN Fundamental Principles of Official Statistics and the NQAF manualQuality dimensions (relevance, accuracy, timeliness, accessibility, coherence)→ STDW3C PROV-O, DCAT, STAC, OGC API and FAIRProvenance, catalogue, geospatial and access standards→
Steward of an index listed here?
Tell us if a description is wrong or out of date; corrections are made in the open.
状态
The Global Risks Index 的方法、结构和定义为预备性工作草案（2026），供审阅和评论，不供实施。在从受治理的证据截取中发布版本之前，不会出现任何数值；届时数值将附有区间、类别、来源及其更正途径。模拟器以您选择的数字运行已发布的方法，绝不显示 NXS-GRX 数值。
NXS-GRX 不是什么
NXS-GRX 是研究和风险分析信息。它不是信用评级或 ESG 评级，不使用任何评级尺度；它不是金融工具的基准，不是官方统计，不是紧急警报或官方分类，也不是建议。不得用于评估自然人。风险等级是已发布尺度上的区间，绝不是对国家或个人的排名。
机构与来源
The Nexus Consortium is stewarded by three independent not-for-profit institutions: The Global Centre for Risk and Innovation (GCRI), through GCRI US in Washington, DC and GCRI Canada in Ontario, with GCRI Canada as Lead Institution, Consortium Secretariat and lead contracting organisation; The Global Risks Forum (GRF); and The Global Risks Alliance (GRA). Technology, infrastructure and service providers work with the Consortium at arm’s length under written agreements, like every other external party in the public-private-planet model, and hold no role in its stewardship.
每个数值均附有其来源的署名，这些来源保留各自的许可。外部名称仅为互操作性而引用。Nexus 撰写的文本拟在发布时采用 CC BY 4.0。
Avis en français
La méthodologie et les définitions du NXS-GRX sont un projet de travail préparatoire (2026). Le NXS-GRX est une information de recherche et d’analyse des risques : ni notation, ni indice de référence financier, ni statistique officielle, ni alerte, ni conseil. Les classes de risque sont des bandes sur une échelle publiée, non des classements.