Banking Nexus · research and knowledge platform; GRA does not lend, invest, insure, rate or advise
Banking Nexus: domain, methods and instruments
Bank risk management is how banks identify, measure, limit and carry the risks they take, from credit and liquidity to operational, model and climate risk, under prudential rules such as Basel III. Each instrument below is illustrative, educational and non-binding.
Domain definition
What Banking Nexus studies
Bank risk management is how banks identify, measure, limit and carry the risks they take, from credit and liquidity to operational, model and climate risk, under prudential rules such as Basel III. Banking Nexus, The Global Risks Alliance's platform for the sector, explains how those risks reach households, firms and the financial system. It assesses no bank and gives no advice.
Banking Nexus studies how banks take, measure and carry risk, and how that risk reaches households, firms and the wider financial system. Its pages cover credit and counterparty exposure, capital and liquidity, operational and third-party resilience, payment dependencies, climate and nature risk, models and AI, governance and Islamic banking. Each page sets out mechanisms, indicators, sources, methods and the board and supervisory decisions the evidence serves. It is shared industry understanding from The Global Risks Alliance, not investment or financial advice.
| 术语 | 定义 |
|---|---|
| Basel III | The Basel Committee's international framework of capital, leverage and liquidity standards for internationally active banks. |
| CET1 比率 | Common equity tier 1 capital as a share of risk-weighted assets, the core measure of a bank's loss-absorbing capital. |
| 流动性覆盖率 | High-quality liquid assets relative to net cash outflows over a thirty-day stress period. |
| Expected credit loss | The probability-weighted estimate of credit losses that IFRS 9 requires banks to provision for. |
Financial system context
The parts of the financial system it covers
-
信用与交易对手风险
贷款账簿和交易风险暴露中的违约、迁移和集中度,以及损失如何在交易对手之间传递。
-
资本与流动性
损失吸收资本、流动性缓冲和融资结构,以及它们在市场和存款人承压时的表现。
-
运营韧性
在网络事件、第三方故障和流程中断期间,于既定影响容忍度以内保持重要服务的连续性。
-
支付与市场基础设施
Payment systems, central counterparties, settlement systems and the digital services banks depend on to move money and collateral.
-
气候与自然风险
Physical and transition risk to loan books and collateral across short, medium and long horizons, explored through scenario analysis.
-
模型与 AI
信用评分、拨备和资本模型,以及机器学习现在所要求的验证和治理。
-
银行业与实体经济
对家庭、小企业、住房和基本服务的贷款,以及银行压力如何变成信贷紧缩。
-
治理与监督
董事会、风险文化、防线、监管审查和处置规划,包括伊斯兰银行业的结构。
风险分类体系
Risks, topic by topic
-
信用与交易对手风险
信用风险是指借款人或交易对手未能按时足额付款的可能性,它仍是大多数银行的主要损失来源。
-
资本与流动性
资本吸收损失,流动性则使银行能够按期履行其义务。
-
银行运营韧性
运营韧性是银行在中断(无论是网络攻击、技术变更失败、第三方中断还是大流行)中持续提供关键服务的能力。
-
Climate and nature bank risk
气候变化和自然丧失通过借款人、抵押品以及银行所依赖的市场影响银行。
-
支付依赖关系
每家银行都依赖支付系统、中央对手方、证券结算系统、代理行和报文网络来转移资金和抵押品。
-
Model and ai risk
银行依靠模型发放贷款、计提拨备、为产品定价、计算资本和发现欺诈,而且越来越多的模型使用机器学习。
-
Banking and real economy
银行将存款转化为面向家庭、中小企业以及基本服务提供方的贷款。
-
银行治理
许多银行倒闭都可以追溯到治理问题:董事会不了解所承担的风险,激励机制奖励增长而非审慎,或控制职能缺乏提出质疑的地位。
-
Housing property finance
Mortgages and property lending are among the largest exposures on many bank balance sheets, and sharp falls in real estate prices have featured in many banking crises.
-
伊斯兰银行业务
伊斯兰银行在旨在符合伊斯兰商法的合同和治理结构下运营,使用基于销售、租赁和合伙的工具取代计息贷款。
Risk-transmission mechanisms
How these risks develop and spread
| 主题 | 机制 | 运作方式 |
|---|---|---|
| 信用与交易对手风险 | 质量在违约之前就已下降 | Arrears, covenant breaches, rating downgrades and forbearance usually precede default by several quarters. Under IFRS 9 a significant increase in credit risk moves a loan from a twelve-month to a lifetime loss allowance, so provisions rise before losses are realised. |
| 信用与交易对手风险 | Losses concentrate | Exposures cluster in sectors, regions, collateral types and large single names. A shock to one segment, such as commercial property or an export industry, can drive losses well beyond what a diversified portfolio model implies. |
| 资本与流动性 | 损失消耗资本 | Credit, market and operational losses reduce common equity. As capital ratios approach buffer levels, restrictions on distributions apply and banks may cut lending to rebuild ratios, passing the stress on to borrowers. |
| 资本与流动性 | 资金可以快速到位 | 无保险存款、批发融资和有担保融资可在数日内被撤出,在数字银行和信息快速传播的情况下更快。存款基础集中与期限错配使挤兑自我强化。 |
| 银行运营韧性 | 依赖关系被隐藏 | A single payment or lending service rests on people, processes, data, technology and external providers. Mapping these dependencies end to end often reveals single points of failure that no one clearly owns. |
| 银行运营韧性 | Third parties concentrate risk | Many banks rely on the same cloud, software and data providers. An outage or compromise at one provider can hit many institutions at once, turning an operational event at one firm into a sector-wide one. |
| Climate and nature bank risk | 实体风险冲击抵押品和收入 | 洪水、风暴、高温和干旱会损坏财产并扰乱生产。未投保的损失由借款人承担,从而提高违约概率和违约损失率,而保险退出可能导致整个区域抵押贷款组合重新定价。 |
| Climate and nature bank risk | 转型风险使各行业重新定价 | 碳定价、监管、技术和需求变化可能使资产搁浅,并挤压高排放借款人的利润空间。时机很重要:晚而突然的转型造成的损失比有序转型更为严重。 |
| 支付依赖关系 | 日内需要流动性 | Real-time gross settlement requires banks to fund each payment as it is made. If a large participant delays payments to conserve liquidity, others wait in turn, and gridlock can build within hours. |
| 支付依赖关系 | Central counterparties concentrate risk | 清算对风险暴露进行轧差,并将违约损失共同分担。在压力情形下,各成员的保证金追缴同时上升,成员违约将动用全体成员共有的违约基金。 |
| Model and ai risk | Drift and regime change | Models learn from past data. When the economy, customer behaviour or products change, relationships shift and accuracy decays, often quietly until losses or complaints reveal it. |
| Model and ai risk | 不透明削弱质疑 | Complex machine learning models can be hard to explain to credit officers, customers and supervisors. Effective challenge, a pillar of model governance, is weaker when no one can say why a model decides as it does. |
| Banking and real economy | Credit supply tightens | 当资本或资金短缺时,银行会提高贷款标准和利差,削减信贷额度,并偏向更安全的借款人。选择更少的小型企业和低收入家庭首当其冲。 |
| Banking and real economy | Borrower stress feeds back | Higher rates, unemployment or falling incomes raise arrears. Bank losses then reduce capital and lending further, a loop that deepens downturns and is central to macroprudential policy. |
| 银行治理 | 董事会监督设定框架 | The board approves strategy, risk appetite and senior appointments. Without the expertise and information to challenge management, it cannot see when risk-taking drifts beyond the appetite it set. |
| 银行治理 | 激励塑造行为 | 与短期收入挂钩的薪酬会鼓励冒险行为,而其代价往往在日后才显现。递延支付、扣减和追回机制旨在使薪酬与已实现的风险结果相一致。 |
| Housing property finance | Credit and prices reinforce each other | Easier credit raises demand and prices. Higher prices raise collateral values and allow more borrowing. The loop runs in reverse in a downturn, and loan-to-value at origination shapes how much loss a lender then absorbs. |
| Housing property finance | 利率上升下的可负担性 | Variable-rate and short-fixation mortgages pass rate rises to borrowers quickly. Debt service burdens rise, arrears follow with a lag, and refinancing can become impossible for stretched borrowers. |
| 伊斯兰银行业务 | Contract structure shapes risk | Murabaha sale, ijarah lease, musharakah and mudarabah partnerships and sukuk carry different mixes of credit, market, equity and operational risk. The risk profile depends on the contract's actual terms, not on its label. |
| 伊斯兰银行业务 | Investment accounts share returns | 利润分享投资账户持有人原则上承担损失。竞争压力可能促使银行平滑收益并自行吸收损失,这种转移称为置换商业风险。 |
研究问题
Who decides, and what they need to know
| Decision holder | 提问 | Evidence the platform's pages describe |
|---|---|---|
| 银行董事会和董事会风险委员会 | What credit risk appetite to set, and which sector and single-name limits to impose | 按细分领域划分的集中度指标、阶段迁移和压力损失 |
| 银行董事会和资产负债委员会 | 应持有何种资本目标和管理缓冲,以及如何安排融资结构 | 内部情景和监管情景下的资本与流动性预测 |
| 银行董事会 | Which business services are important and what impact tolerance each should carry | Service maps, scenario test results and incident history |
| 银行董事会和董事会风险委员会 | How climate and nature risks enter risk appetite, strategy and transition planning | Materiality assessment, scenario results and sector exposures |
| 央行兼任监督者与运营者 | Which systems are systemically important and what standards and recovery plans they must meet | Availability, incident and participation data; concentration of flows |
| Bank board and model risk committee | What model risk appetite to set and which uses of AI are acceptable | Model inventory, tiering and validation findings |
| 宏观审慎当局或金融政策委员会 | 是否设定或释放资本缓冲和基于借款人的措施 | 信贷缺口、偿债比率和贷款调查 |
| Bank board and nomination committee | 董事会的技能、独立性和信息是否与银行的风险状况相适应 | Board effectiveness reviews, skills matrices and risk reports |
| 宏观审慎主管机构 | Whether to set or adjust loan-to-value, debt-service-to-income or sectoral capital measures | 价格收入比、信贷增长、放贷标准与拖欠 |
| 银行董事会 | What risk appetite, return smoothing policy and reserve practice to adopt for investment accounts | Displaced commercial risk analysis, return history and capital impact |
These are the decisions of the named institutions. The platform describes the evidence they use; it takes none of these decisions and advises on none.
Key indicators
What is measured, in which unit, by whom
| 指标 | 单位 | 节奏 | 来源类型 | 证据标签 |
|---|---|---|---|---|
| 不良贷款率 | 占贷款总额的百分比 | 每季度 | Supervisory returns, Pillar 3 disclosures | 已发布统计 |
| 第 2 阶段和第 3 阶段贷款占比 | 占贷款总额的百分比 | 每季度 | IFRS 9 财务报表 | 已报告 |
| Common equity tier 1 ratio | 占风险加权资产的百分比 | 每季度 | Supervisory returns, Pillar 3 disclosures | 已发布统计 |
| 杠杆率 | 占总风险暴露度量的百分比 | 每季度 | 监管报表 | 已发布统计 |
| Important business services within impact tolerance in testing | 已测试的服务数量及通过数量 | 年度 | 情景测试记录 | 已报告 |
| Major ICT-related incidents | 按严重程度类别统计的数量 | 每季度 | 向监管机构报送的事件报告 | 已发布统计 |
| 融资排放 | 二氧化碳当量吨 | 年度 | 适用于贷款和投资账簿的 PCAF 方法 | 估算 |
| 对气候政策相关行业的风险暴露 | percent of loan book | 年度 | Pillar 3 ESG disclosures | 已报告 |
| Value settled through large-value payment systems | 每天的货币单位数 | 每日 | System operator statistics | 已发布统计 |
| 支付系统可用性 | percent of operating hours | 每月 | 运营方披露 | 已报告 |
| Model inventory coverage | count of models by tier and validation status | 每季度 | Model risk management records | 已报告 |
| 信用评分卡的区分能力 | 基尼系数或曲线下面积 | 年度 | 模型验证 | 模型估算 |
| 信贷与 GDP 之比缺口 | 百分点 | 每季度 | 国际清算银行(BIS)信贷统计 | 估算 |
| 企业贷款的信贷标准 | 报告收紧的银行净占比 | 每季度 | 银行贷款调查 | 已报告 |
| Board members with banking or risk expertise | 占董事会比例 | 年度 | 治理披露 | 已声明 |
| 独立董事 | 占董事会比例 | 年度 | 治理披露 | 已声明 |
| 发放时的贷款价值比 | 占财产价值的百分比 | 每季度 | 贷款层面数据和监管数据 | 已发布统计 |
| 发放时的债务偿还收入比 | percent of gross income | 每季度 | Loan-level data, household surveys | 已报告 |
| 伊斯兰银行的资本充足率 | 占风险加权资产的百分比 | 每季度 | IFSB 审慎指标 | 已发布统计 |
| Non-performing financing ratio | 占融资总额的百分比 | 每季度 | IFSB 审慎指标 | 已发布统计 |
Indicators carry a name, a unit and the kind of evidence behind them. This page publishes no value for them: values belong to their sources, with their own dates and definitions.
方法论
The methods the platform's work rests on
-
金融经济学
金融经济学与宏观金融解释银行资产负债表、信贷和资产价格如何与产出、就业和公共财政相互作用。在银行风险中,它们将情景与损失和资本联系起来,说明银行压力如何反馈至经济,并使宏观审慎和货币决策建立在经检验的关系之上。
-
Probability statistics
Probability and statistics underlie every credit, market and operational risk measure in banking. Default probabilities, loss distributions, value at risk and stress losses all rest on estimation from limited and often unrepresentative data. The discipline makes the uncertainty in those estimates explicit for the people who rely on them.
-
网络科学
Network science maps how banks are linked through interbank lending, derivatives, common asset holdings, payment flows and shared service providers. It shows where a single failure could cascade, which institutions are central to the system, and why contagion differs from the sum of individual bank risks.
-
Financial engineering
Financial and risk engineering supplies the instruments and models banks use to measure, price, hedge and transfer risk, from expected credit loss and counterparty exposure to value at risk and capital allocation. It also supplies the discipline of model risk management that keeps those tools reliable.
-
Ai ml
Artificial intelligence and machine learning now support credit underwriting, fraud and financial crime detection, early warning of borrower stress and supervisory analysis. In banking they are both a tool and a source of risk: their outputs must be explainable, fair, robust and validated before they influence consequential decisions about customers or capital.
数据来源
证据的来源
| 来源 | 它提供什么 | 许可类别 |
|---|---|---|
| 国际货币基金组织金融稳健指标 | 各国主管部门报告的不良贷款、拨备和行业贷款分布 | 免费,适用 IMF 使用条款 |
| EBA risk dashboard and EU-wide transparency exercise | 欧盟银行层面的资产质量、宽限措施和国际财务报告准则第 9 号阶段数据 | 免费,须署名 |
| ECB Supervisory Banking Statistics | Asset quality and credit risk aggregates for significant institutions in the euro area | 免费,须署名 |
| FFIEC Call Reports and Federal Reserve FR Y-9C | Loan categories, past-due and charge-off data for US banks and holding companies | 公有领域 |
| BIS consolidated banking statistics | Banks' foreign claims by counterparty country and sector on an ultimate-risk basis | 免费,须遵守 BIS 使用条款 |
| Rating agency default and recovery studies | Historical default and recovery rates by rating grade and sector | 许可授权,订阅 |
| 国家中央信贷登记系统 | Borrower-level exposures for concentration and interconnection analysis | Restricted, supervisory access |
| BIS 巴塞尔 III 监测报告 | 国际活跃银行的资本、杠杆和流动性汇总头寸 | 免费,须遵守 BIS 使用条款 |
| Federal Reserve supervisory stress test results | Scenario design, projected losses and post-stress capital for large US banks | 公有领域 |
| Bank of England stress testing results | 英国主要银行同步压力测试和探索性压力测试的情景设计与结果 | 免费,英格兰银行使用条款 |
| EBA 欧盟范围压力测试 | Bank-level stressed capital projections under a common methodology | 免费,须署名 |
| 第 3 支柱披露 | 银行公布的资本构成、风险加权资产、LCR 和 NSFR | 公开,银行条款 |
| Registers and critical provider designations under the EU Digital Operational Resilience Act | 主管机构持有的信息通信技术第三方安排登记册,以及已指定关键信息通信技术服务提供方的公开名单 | 登记册仅限主管部门查阅;指定名单公开 |
| ECB cyber resilience stress test and supervisory publications | 关于网络响应和恢复能力的监管发现 | 免费,须署名 |
| Bank of England and PRA operational resilience publications | Policy statements, thematic findings and CBEST threat-led testing reports | 免费,英格兰银行使用条款 |
| ORX operational risk loss data | Pooled loss events by event type and business line from member banks | Licensed, membership |
| ENISA 金融部门威胁态势 | Analysis of cyber threats and incident patterns affecting financial institutions | 免费,须署名 |
| Pillar 3 operational risk disclosures | Operational risk capital and loss summaries published by banks | 公开,银行条款 |
| NGFS 情景 | Transition and physical risk pathways with macroeconomic variables for scenario analysis | 免费,须署名 |
| ECB climate stress test and climate-related statistical indicators | Supervisory exercise methodology and experimental indicators of euro area bank exposures | 免费,须署名 |
| Bank of England Climate Biennial Exploratory Scenario | 针对英国银行和保险公司的探索性演练的设计与已发布结果 | 免费,英格兰银行使用条款 |
| 依据 EBA 标准的第 3 支柱 ESG 披露 | 银行报告的对高碳强度行业的风险暴露、实体风险和分类标准一致性 | 公开,银行条款 |
| PCAF Global GHG Accounting and Reporting Standard | Methodology and data quality scores for financed emissions | 标准免费;排放因子数据库按会员资格提供 |
| Copernicus Climate Change Service and JRC hazard maps | 针对洪水、高温和干旱的物理危害图层 | 免费开放,须署名 |
Instruments explained for research
Analytical instruments
Each instrument runs a named, cited method on stated teaching values, shows its uncertainty and limits, and keeps its data in a table. All of them are illustrative, educational and non-binding: none quotes a premium, a rate, a price, a rating or an eligibility, and none describes a real institution, portfolio or place.
信用风险
A credit portfolio's loss distribution
Illustrative, educational and non-binding
The one-factor model behind risk-based capital: the expected loss is probability of default times loss given default; correlation between borrowers spreads losses into rare bad years. The gap between the 99.9 % loss and the expected loss is the unexpected loss that capital is meant to absorb.
- 投资组合损失
- Expected loss
With low correlation the curve stays close to the expected loss; with high correlation bad years cluster, and the tail rises steeply.
控制措施
- Probability of default
- 1.00 % a year
- 违约损失率
- 45 % of exposure
- Asset correlation
- 0.12
结果
- Expected loss (% of exposure a year)
- 0.45
- Loss at 99 % (% of exposure)
- 2.36
- Loss at 99.9 % (% of exposure)
- 4.06
- Unexpected loss at 99.9 % (% of exposure)
- 3.61
Data, source and model record
| 重现期(年) | Loss (% of exposure) |
|---|---|
| 2 | 0.30 |
| 4 | 0.59 |
| 8 | 0.91 |
| 17 | 1.28 |
| 34 | 1.68 |
| 70 | 2.13 |
| 141 | 2.60 |
| 288 | 3.11 |
| 585 | 3.64 |
| 1,189 | 4.20 |
| 2,418 | 4.79 |
| 4,918 | 5.41 |
| 10,000 | 6.04 |
- 来源
- This page's one-factor model (the network's own records). Methods: Asymptotic single risk factor (one-factor) credit loss distribution (Vasicek (2002), The distribution of loan portfolio value, Risk 15(12)); The model behind the Basel IRB risk-weight functions (Basel Committee on Banking Supervision (2005), An Explanatory Note on the Basel II IRB Risk Weight Functions).
- 日期
- Instrument record 1.0.0, 2026-11-16; teaching values chosen inside the ranges the cited note discusses.
- 范围
- A hypothetical loan portfolio of a hypothetical bank. It names no bank, borrower, sector or country, and represents none.
- 单位
- Losses in percent of total exposure; return periods in years (1 in 1,000 years is the 99.9 % level).
- Currency
- None: every value is a share (percent), so no currency or exchange rate applies
- 状态
- Illustrative, educational and non-binding: no premium, rate, price, rating, eligibility or advice is given or implied.
Model record
- 建模
- gra.credit.one-factor
- 版本
- 1.0.0
- 负责人
- GRA platforms lane
- 目的
- To show how probability of default, loss given default and correlation between borrowers shape a portfolio's loss distribution: expected loss, high quantiles and unexpected loss.
- Risk domain
- 信用与交易对手风险
- 假设
- Every borrower has the same probability of default, loss given default and asset correlation. Defaults depend on one common factor and an independent part (one-factor Gaussian model). The portfolio is infinitely granular unless a large exposure is set.
- 输入
- Probability of default (% a year); Loss given default (% of exposure); Asset correlation
- 数据
- None: teaching values set by the controls; no bank's data.
- 方法
- Asymptotic single risk factor (one-factor) credit loss distribution (Vasicek (2002), The distribution of loan portfolio value, Risk 15(12)); The model behind the Basel IRB risk-weight functions (Basel Committee on Banking Supervision (2005), An Explanatory Note on the Basel II IRB Risk Weight Functions)
- 验证
- Checked against the closed form of the model, including the 99.9 % conditional default probability of the IRB formula (the network's own records). Not calibrated to any portfolio.
- 局限性
- One factor, homogeneous borrowers, no migration, no maturity adjustment, no collateral dynamics. A capital concept: it is not the regulatory capital requirement of any bank or jurisdiction.
- 不确定性
- Parameter uncertainty is not shown: move the controls to see how far the tail moves with probability of default and correlation.
- Approved use
- Education and research on credit loss distributions and the ideas behind risk-based capital.
- Prohibited use
- Credit decisions, ratings, regulatory capital calculation, provisioning, or any statement about a real bank or borrower.
- Runs on
- Static: computed at build time and in the reader's browser; no server, no data leaves the page.
集中度
One large exposure and the tail
Illustrative, educational and non-binding
The same portfolio with one borrower holding a share of it. The expected loss does not change; the tail does, because the large exposure's default is a lump the rest of the portfolio cannot average away. The Herfindahl index and the effective number of exposures summarise it.
- Granular portfolio
- With the large exposure
Move the large exposure from zero: the 99.9 % loss rises above the granular curve, by more than its share times the loss given default would suggest at low default probabilities.
控制措施
- Share of the largest exposure
- 5.0 % of the portfolio
- Probability of default
- 1.00 % a year
- 违约损失率
- 45 % of exposure
- Asset correlation
- 0.12
结果
- Loss at 99.9 %, granular portfolio (% of exposure)
- 4.06
- Loss at 99.9 %, with the large exposure (% of exposure)
- 4.55
- Concentration add-on (% of exposure)
- 0.49
- Effective number of exposures (1 / HHI of the large exposure)
- 400
Data, source and model record
| 重现期(年) | Granular (%) | With the large exposure (%) |
|---|---|---|
| 2 | 0.30 | 0.28 |
| 4 | 0.59 | 0.56 |
| 8 | 0.91 | 0.89 |
| 17 | 1.28 | 1.29 |
| 34 | 1.68 | 1.81 |
| 70 | 2.13 | 2.53 |
| 141 | 2.60 | 3.06 |
| 288 | 3.11 | 3.57 |
| 585 | 3.64 | 4.12 |
| 1,189 | 4.20 | 4.69 |
| 2,418 | 4.79 | 5.29 |
| 4,918 | 5.41 | 5.91 |
| 10,000 | 6.04 | 6.56 |
- 来源
- This page's model: the one-factor distribution with one exposure added exactly, integrated over the common factor (the network's own records). Methods: Asymptotic single risk factor (one-factor) credit loss distribution (Vasicek (2002), The distribution of loan portfolio value, Risk 15(12)); The model behind the Basel IRB risk-weight functions (Basel Committee on Banking Supervision (2005), An Explanatory Note on the Basel II IRB Risk Weight Functions); Credit risk concentration: name and sector concentration (Basel Committee on Banking Supervision (2006), Studies on credit risk concentration, Working Paper 15).
- 日期
- Instrument record 1.0.0, 2026-11-16.
- 范围
- A hypothetical loan portfolio of a hypothetical bank. It names no bank, borrower, sector or country, and represents none.
- 单位
- Losses in percent of total exposure; return periods in years (1 in 1,000 years is the 99.9 % level).
- Currency
- None: every value is a share (percent), so no currency or exchange rate applies
- 状态
- Illustrative, educational and non-binding: no premium, rate, price, rating, eligibility or advice is given or implied.
Model record
- 建模
- gra.credit.concentration
- 版本
- 1.0.0
- 负责人
- GRA platforms lane
- 目的
- To show why single-name concentration raises tail loss while leaving expected loss unchanged.
- Risk domain
- 信用与交易对手风险
- 假设
- Every borrower has the same probability of default, loss given default and asset correlation. Defaults depend on one common factor and an independent part (one-factor Gaussian model). The portfolio is infinitely granular unless a large exposure is set.
- 输入
- Share of the largest exposure (%); Probability of default (% a year); Loss given default (% of exposure); Asset correlation
- 数据
- None: teaching values set by the controls; no bank's data.
- 方法
- Asymptotic single risk factor (one-factor) credit loss distribution (Vasicek (2002), The distribution of loan portfolio value, Risk 15(12)); The model behind the Basel IRB risk-weight functions (Basel Committee on Banking Supervision (2005), An Explanatory Note on the Basel II IRB Risk Weight Functions); Credit risk concentration: name and sector concentration (Basel Committee on Banking Supervision (2006), Studies on credit risk concentration, Working Paper 15)
- 验证
- Checked against the closed form of the model, including the 99.9 % conditional default probability of the IRB formula (the network's own records). Not calibrated to any portfolio.
- 局限性
- One factor, homogeneous borrowers, no migration, no maturity adjustment, no collateral dynamics. A capital concept: it is not the regulatory capital requirement of any bank or jurisdiction.
- 不确定性
- Parameter uncertainty is not shown: move the controls to see how far the tail moves with probability of default and correlation.
- Approved use
- Education and research on credit loss distributions and the ideas behind risk-based capital.
- Prohibited use
- Credit decisions, ratings, regulatory capital calculation, provisioning, or any statement about a real bank or borrower. Setting or assessing large-exposure limits.
- Runs on
- Static: computed at build time and in the reader's browser; no server, no data leaves the page.
流动性
The structure of the liquidity coverage ratio
Illustrative, educational and non-binding
A stylised balance sheet run through the Basel III liquidity coverage ratio's arithmetic: each class of funding runs off at the standard's baseline rate over a 30-day stress, inflows count up to 75 % of outflows, and the stock of high-quality liquid assets is set against the net outflow. The stress control scales every run-off rate.
- Outflow over 30 days by funding class
- Net outflows after capped inflows
- Stock of high-quality liquid assets
Funding from financial institutions runs off in full, stable retail deposits barely at all: the mix of funding matters as much as the size of the buffer.
控制措施
- High-quality liquid assets
- 15.0 % of total funding
- Stable retail deposits
- 30 % of total funding
- Less stable retail deposits
- 20 % of total funding
- Operational deposits
- 15 % of total funding
- Non-operational deposits of non-financial corporates
- 15 % of total funding
- Unsecured funding from financial institutions
- 10 % of total funding
- Contractual inflows over 30 days
- 5.0 % of total funding
- Stress on run-off rates
- 1.00 times the baseline
结果
- Outflows over 30 days (% of total funding)
- 23.3
- Inflows counted (capped at 75 % of outflows) (% of total funding)
- 5.00
- Net outflows (% of total funding)
- 18.3
- Liquid assets over net outflows (%)
- 82
Data, source and model record
| Funding class | Balance (% of funding) | Run-off rate (%) | Outflow (% of funding) |
|---|---|---|---|
| Stable retail deposits | 30.0 | 5 | 1.50 |
| Less stable retail deposits | 20.0 | 10 | 2.00 |
| Operational deposits | 15.0 | 25 | 3.75 |
| Non-operational deposits of non-financial corporates | 15.0 | 40 | 6.00 |
| Unsecured funding from financial institutions | 10.0 | 100 | 10.0 |
- 来源
- Run-off rates and the 75 % inflow cap: Basel Committee on Banking Supervision (2013), Basel III: The Liquidity Coverage Ratio and liquidity risk monitoring tools (BCBS 238), baseline rates for the classes shown (Stable retail deposits 5 %; Less stable retail deposits 10 %; Operational deposits 25 %; Non-operational deposits of non-financial corporates 40 %; Unsecured funding from financial institutions 100 %). Arithmetic: this page's model (the network's own records).
- 日期
- BCBS 238 published January 2013; instrument record 1.0.0, 2026-11-16. Jurisdictions implement and adjust these rates: check the rules that apply before relying on any rate.
- 范围
- A hypothetical bank's funding mix. The listed classes need not add up to 100 %: the rest is equity and funding that does not fall due within 30 days.
- 单位
- Percent of total funding; the ratio in percent.
- Currency
- None: every value is a share (percent), so no currency or exchange rate applies
- 状态
- Illustrative, educational and non-binding: no premium, rate, price, rating, eligibility or advice is given or implied.
Model record
- 建模
- gra.liquidity.lcr-structure
- 版本
- 1.0.0
- 负责人
- GRA platforms lane
- 目的
- To show how the funding mix, run-off rates, the inflow cap and liquid assets combine in the liquidity coverage ratio.
- Risk domain
- 流动性风险
- 假设
- Five funding classes at the standard's baseline run-off rates. All liquid assets count at full value (no haircuts, no caps by asset level). Inflows are a single contractual amount, capped at 75 % of outflows.
- 输入
- High-quality liquid assets and funding balances (% of total funding); Contractual inflows (% of total funding); Stress multiplier on run-off rates
- 数据
- BCBS 238 (January 2013): baseline run-off rates and the inflow cap.
- 方法
- Liquidity coverage ratio (Basel Committee on Banking Supervision (2013), BCBS 238)
- 验证
- Checked against hand-computed outflows and the inflow cap (the network's own records).
- 局限性
- Omits asset haircuts and level caps, secured funding, committed facilities, derivatives and every jurisdictional option. It is not an LCR calculation, and its result says nothing about compliance.
- 不确定性
- None modelled: the stress control shows how far the ratio moves when run-off rates are worse than the baseline.
- Approved use
- Education on the structure of the liquidity coverage ratio.
- Prohibited use
- Calculating, reporting or assessing any bank's LCR or compliance; supervisory or funding decisions.
- Runs on
- Static: computed at build time and in the reader's browser; no server.
风险暴露网络
资产冲击如何演变为系统性损失
Illustrative, educational and non-binding
Banks hold claims on each other. A fall in the value of their external assets distresses each bank in proportion to its leverage, and DebtRank passes that distress along interbank exposures, round by round, until it settles. The amplification is the system loss beyond the initial hit.
- System equity lost
Small shocks are amplified by the network; past a point, banks default in the first round and there is little left to amplify.
控制措施
- 外部资产下降
- 2.0 %
结果
- 外部资产下降 (%)
- 2.0
- System equity lost in the first round (%)
- 18.0
- System equity lost after contagion (%)
- 45.0
- Amplification by the network (percentage points)
- 27.0
- Banks in default (of 20)
- 0
Data, source and model record
| 轮次 | 系统权益损失(%) | 违约银行 |
|---|---|---|
| 0 | 18.0 | 0 |
| 1 | 28.8 | 0 |
| 2 | 35.3 | 0 |
| 3 | 39.2 | 0 |
| 4 | 41.5 | 0 |
| 5 | 42.9 | 0 |
| 6 | 43.7 | 0 |
| 7 | 44.2 | 0 |
| 8 | 44.5 | 0 |
| 9 | 44.7 | 0 |
| 10 | 44.8 | 0 |
| 11 | 44.9 | 0 |
| 12 | 44.9 | 0 |
| 13 | 45.0 | 0 |
| 14 | 45.0 | 0 |
| 15 | 45.0 | 0 |
| 16 | 45.0 | 0 |
| 17 | 45.0 | 0 |
| 18 | 45.0 | 0 |
| 19 | 45.0 | 0 |
| 20 | 45.0 | 0 |
| 21 | 45.0 | 0 |
| 22 | 45.0 | 0 |
| 23 | 45.0 | 0 |
| 24 | 45.0 | 0 |
| 25 | 45.0 | 0 |
| 26 | 45.0 | 0 |
| 27 | 45.0 | 0 |
| 28 | 45.0 | 0 |
| 29 | 45.0 | 0 |
| 30 | 45.0 | 0 |
- 来源
- Computed by the estate's debtrank model (the network's own records, sha256 e8f84afa4822), the same code the platform's existing simulators run. Methods: DebtRank (Battiston and others (2012), Scientific Reports 2:541); Differential DebtRank (Bardoscia and others (2015), PLOS ONE 10); Contagion in financial networks (Gai and Kapadia (2010), Proc. R. Soc. A 466); Clearing in financial networks (Eisenberg and Noe (2001), Management Science 47); Uncertainty fan (parameter Monte Carlo) (Spiegelhalter, Pearson and Short (2011), Visualizing uncertainty about the future, Science 333); One-at-a-time sensitivity (tornado) (Saltelli and others (2008), Global Sensitivity Analysis: The Primer, Wiley). Scenarios: the catalogue's teaching values (leverage 10 times, interbank share 10 %, recovery 40 %, 20 banks, seed 62036) with the fall in external assets varied.
- 日期
- Instrument record 1.0.0, 2026-11-16; teaching values from the estate's instrument catalogue (the network's own records).
- 范围
- A generated network of hypothetical banks (a core and a periphery). It names no bank or banking system, and represents none.
- 单位
- Percent of total system equity; rounds of contagion.
- Currency
- None: every value is a share (percent), so no currency or exchange rate applies
- 状态
- Illustrative, educational and non-binding: no premium, rate, price, rating, eligibility or advice is given or implied.
Model record
- 建模
- gra.network.debtrank
- 版本
- 1.0.0
- 负责人
- GRA platforms lane (model code: the estate's instrument engine, W13)
- 目的
- To show how interconnection and leverage amplify a common shock in a banking network.
- Risk domain
- Systemic and contagion risk
- 假设
- Distress passes linearly along interbank exposures (DebtRank), each node once per round. External assets fall by the same share for every bank. Network: a densely linked core and a sparsely linked periphery, generated from the seed.
- 输入
- Fall in external assets (%); Leverage (assets over equity) (times); Interbank share of assets (%); Recovery on interbank claims (%); Banks in the system (count); Core banks (% of banks)
- 数据
- None: a generated network with teaching values; no supervisory or market data.
- 方法
- DebtRank (Battiston and others (2012), Scientific Reports 2:541); Differential DebtRank (Bardoscia and others (2015), PLOS ONE 10); Contagion in financial networks (Gai and Kapadia (2010), Proc. R. Soc. A 466); Clearing in financial networks (Eisenberg and Noe (2001), Management Science 47); Uncertainty fan (parameter Monte Carlo) (Spiegelhalter, Pearson and Short (2011), Visualizing uncertainty about the future, Science 333); One-at-a-time sensitivity (tornado) (Saltelli and others (2008), Global Sensitivity Analysis: The Primer, Wiley)
- 验证
- The engine's code, run unchanged at build time (the network's own records, checked current by its --check).
- 局限性
- No fire sales, no funding runs, no clearing (Eisenberg and Noe) or recapitalisation, one generated network. Results depend on the network drawn.
- 不确定性
- Not shown on this page; the engine's sensitivity of the final loss is listed in the table below the chart.
- Approved use
- Education and research on contagion mechanisms and macroprudential stress testing ideas.
- Prohibited use
- Statements about real banks or systems; supervisory, investment or credit decisions.
- Runs on
- Static: computed at build time; the reader's browser only switches between precomputed scenarios.
情景分析
Scenarios, not forecasts
A scenario is a set of named assumptions about the future, run through a model to see what follows; it is not a prediction. The instruments above move one assumption at a time; scenario work moves several together, with every assumption stated.
Regulatory context and related standards
The frameworks this work is read against
| 框架 | 机构 | 它规定了什么 | Edition or date | Where it applies |
|---|---|---|---|---|
| 巴塞尔 III 框架 | 巴塞尔银行监管委员会 | 最低资本、杠杆、流动性覆盖和稳定资金要求,以及在此之上的资本缓冲 | December 2010, finalised December 2017; consolidated in the Basel Framework | Basel Committee members, by national rules on their own timetables |
| 有效银行监管核心原则 | 巴塞尔银行监管委员会 | 关于监管权力、许可、风险管理和治理的基准标准,在国际货币基金组织和世界银行的金融部门评估规划(FSAP)中进行评估 | Revised April 2024 | International standard for supervisors; assessed in IMF and World Bank FSAPs |
| 有效管理和监督气候相关金融风险的原则 | 巴塞尔银行监管委员会 | 对银行治理、气候相关风险管理和情景分析及其监督审查的期望 | June 2022 | Basel Committee members, through national supervisors |
| 金融机构有效处置机制的关键属性 | 金融稳定理事会 | 处置权力、恢复与处置计划以及跨境合作,使银行能够在无需纳税人支持的情况下倒闭 | October 2011, updated October 2014 | FSB member jurisdictions |
| 国际财务报告准则第 9 号:金融工具 | 国际会计准则理事会 | 金融资产的分类与三阶段前瞻性预期信用损失计提 | Issued July 2014; effective 1 January 2018 | Jurisdictions that apply IFRS (US GAAP uses CECL instead) |
| NGFS 气候情景 | 央行与监管机构绿色金融网络 | 央行和监管机构用于气候情景分析的参考转型和实体风险路径 | Phase V, November 2024 | Reference scenarios for NGFS members; not binding |
| IFRS S2 气候相关披露 | 国际可持续准则理事会 | 气候相关风险和机遇的披露,包括融资排放和气候韧性 | Issued June 2023, amended December 2025 | Where a jurisdiction adopts it |
| 监管审查与评估流程(SREP) | 欧洲银行管理局和欧洲央行银行监管 | 评估商业模式、治理以及资本和流动性风险的监管流程,据此设定第 2 支柱要求和指引 | EBA Guidelines EBA/GL/2022/03 (March 2022) | European Union; the ECB for significant euro area banks |
| 综合资本分析与审查(CCAR)和监管压力测试 | 美国联邦储备系统理事会 | 针对美国大型银行的压力情景和资本计划审查监管流程,据此设定压力资本缓冲 | Annual; capital plan rule with the stress capital buffer since 2020 | United States, large bank holding companies |
| ISO 31000 风险管理指南 | 国际标准化组织 | 在整个组织内管理风险的原则、框架和流程 | ISO 31000:2018 | International, voluntary |
Frameworks change and apply differently by jurisdiction. Check the current version and its applicability with the issuing body before relying on it; nothing here is legal or compliance guidance, and no comparison here is a supervisory determination.
Institutional counterparts
The institutions this work speaks to
- 银行董事会和董事会风险委员会
- Chief risk officer and credit committee
- 审慎监管机构
- 宏观审慎主管机构
- 银行董事会和资产负债委员会
- 通过 SREP 或 CCAR 开展工作的审慎监管机构
- 中央银行
- 银行董事会
- 首席运营官和首席信息安全官
- 监督关键第三方的机构
- Credit risk function
- NGFS 中的央行和监管机构
- 央行兼任监督者与运营者
- Bank treasury
- 中央对手方董事会和风险委员会
- Resolution authority
研究出版物
Research, methods and sources
Educational resources
Learn the methods
Readiness workflows
金融准备度
Finance readiness is the structured preparation of evidence for financial-sector analysis and due diligence. It is not financing approval: reaching any readiness stage never makes a project, institution or portfolio eligible for funding, cover or credit.
Contribution mechanisms
参与
边界
What this page is not
Banking Nexus is part of The Global Risks Alliance, which does not accept deposits, extend credit, underwrite or bind insurance, arrange or execute investments, offer securities, broker regulated products, issue credit ratings or give financial-product recommendations. Every instrument on this page is illustrative, educational and non-binding: it quotes no premium, rate, price, rating or eligibility, and it describes no real institution, portfolio or place.