Skip to content

GRA · Banking Nexus

空间金融资本与危害、风险暴露和转型的交汇之处

银行运营韧性

Operational resilience is a bank’s ability to keep delivering critical services through disruption, whether from a cyber attack, a failed technology change, a third-party outage or a pandemic. Supervisors now ask banks to identify their important business services, set impact tolerances and show they can stay within them. This page explains how operational disruption develops and spreads, how resilience is measured and tested, and who decides what level of disruption is tolerable.

01登记

记录在案的银行业务韧性。

02问题

Bank operational resilience: how the risk develops.

第三方、网络与流程连续性。下面的档案说明风险如何发展、衡量什么、数据来自何处、如何建模以及由谁决定。

机制

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.

机制

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.

机制

网络事件不断升级

入侵、勒索软件或数据损坏可能使服务瘫痪并损害记录的完整性。当备份遭到破坏,或攻击通过共享网络和供应链蔓延时,恢复会更为缓慢。

机制

Disruption becomes financial

An outage blocks payments, delays settlement and erodes customer confidence. Prolonged disruption can trigger deposit outflows, sanctions and compensation costs, and at a market utility it can cause knock-on failures elsewhere.

03压力测试

资产冲击如何演变为系统性损失。

银行间传染之后系统损失的权益,相对于最初的冲击,随杠杆、互联程度和回收率变化而变化,这是宏观审慎 压力测试的核心。这是带有明示示意数值的教学模型,不是预测。

控制措施

外部资产下降
2.0%
杠杆率(资产与权益之比)
10 倍
银行间资产占比
10%
银行间债权的回收
40%
体系内的银行
20
核心银行
占银行的 20%

读数

初始冲击造成的权益损失
18.0%
传染后的权益损失
45.0%
传染新增(DebtRank)
27.0 个点
违约银行
0
数字、方法与来源
轮次系统权益损失(%)违约银行
018.000
128.800
235.280
339.170
441.500
542.900
643.740
744.240
844.550
944.730
1044.840
1144.900
1244.940
1344.960
1444.980
1544.990
1644.990
1745.000
1845.000
1945.000
2045.000
2145.000
2245.000
2345.000
2445.000
2545.000
2645.000
2745.000
2845.000
2945.000
3045.000

方法

  1. DebtRank Battiston and others (2012), Scientific Reports 2:541
  2. Differential DebtRank Bardoscia and others (2015), PLOS ONE 10
  3. Contagion in financial networks Gai and Kapadia (2010), Proc. R. Soc. A 466
  4. 金融网络中的清算 Eisenberg 和 Noe(2001),Management Science 47
  5. Uncertainty fan (parameter Monte Carlo) Spiegelhalter, Pearson and Short (2011), Visualising uncertainty about the future, Science 333
  6. One-at-a-time sensitivity (tornado) Saltelli and others (2008), Global Sensitivity Analysis: The Primer, Wiley

数据

No observed data enters this model: it runs on the teaching values set by the controls, to be replaced with a programme’s own evidence.

资产冲击如何演变为系统损失。示意性教学模型:它依据控件设定的数值运行,而不是实测数据。方法:DebtRank(Battiston 等(2012),Scientific Reports 2:541);Differential DebtRank(Bardoscia 等(2015),PLOS ONE 10);金融网络中的传染(Gai 和 Kapadia(2010),Proc. R. Soc. A 466)。
04指标

衡量什么,以及衡量得如何。

每个指标都注明其单位、频率、背后的来源类型以及数值将附带的证据标签。此处不显示数值:数值由计划依据其自身权限提供。

指标单位节奏来源类型标签
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
Service downtimeminutes per incident每个事件事件日志观测
Recovery time against objective小时每个事件Continuity tests and incident records观测
关键第三方集中度每个提供方的重要服务数量年度Register of third-party arrangements已报告
运营风险 损失currency units by Basel event type每季度Internal loss data已报告
超过整改期限的关键漏洞计数每月Security management records观测
05风险暴露、损失与资本

银行运营韧性对风险暴露、损失和资本意味着什么。

风险官、监管者或财政部会要求的风险暴露、损失、情景和资本视图,以及它们据以报告的框架。结构遵循监管和精算实务。数值为说明性:绝非价格、评级或建议。

情景路径

在各国央行和监管机构所用的情景系列下,银行运营韧性可能承受怎样的压力。

有序无序温室世界202520302035204020452050高低
情景系列转型风险物理风险压力测试问什么
有序早期、渐进的政策较低账面能否承受持续的重新定价?
无序迟缓而突然的政策中低在突然重新定价和信用迁移下,什么会出现问题?
温室世界有限政策高且上升实体损失集中在抵押品和保险的哪些环节?
按 NGFS 定义的情景类别;曲线为示意性的综合压力指数,并非 NGFS 变量或预测。
NGFS 气候情景 · 央行与监管机构绿色金融网络IFRS S2 气候相关披露 · ISSBTCFD 建议 · 金融稳定理事会 TCFD有效管理气候相关金融风险的原则 · 巴塞尔银行监管委员会
06来源

证据来自何处。

具名公共数据集与机构记录,各自注明许可类别。依协议获得的来源仅按其条款使用,绝不再次发布。

来源它提供什么许可类别
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
欧洲央行网络韧性 压力测试 和监管出版物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 publicationsPolicy statements, thematic findings and CBEST threat-led testing reports免费,英格兰银行使用条款
ORX 操作风险 损失数据Pooled loss events by event type and business line from member banksLicensed, 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 disclosuresOperational risk capital and loss summaries published by banks公开,银行条款
07决定

由谁决定,以及证据服务于什么。

Banking Nexus 负责准备证据。以下决定由所列机关依其自身规则作出,绝不由平台作出。

谁来决定问题本页面服务的证据
银行董事会Which business services are important and what impact tolerance each should carryService maps, scenario test results and incident history
首席运营官和首席信息安全官在何处投资于第三方的冗余、恢复和退出计划依赖关系图、恢复测试、脆弱性和事件数据
审慎监管机构Whether a bank’s self-assessment and testing are credible, and when to interveneIncident reports, test evidence and thematic reviews
监督关键第三方的机构Which providers to designate and what oversight to applyConcentration across the sector and the substitutability of each provider

该主题产生的产出

  • Guides to mapping important business services and their dependencies
  • 用于桌面推演和实战测试的严重但合理的中断情景
  • 关于云和关键技术提供方集中度的研究
  • 关于事件模式的简报,附来源和证据标签
08核心问题

行业提出的共同问题。

  1. 01

    Which services matter most?

    Which business services, if disrupted, would harm customers or market integrity.

  2. 02

    What do they depend on?

    每项服务背后的人员、流程、技术、设施和第三方。

  3. 03

    共同依赖在哪里?

    Providers and systems that many banks rely on at once.

  4. 04

    服务能以多快的速度恢复?

    银行在严峻情景下能否保持在影响容忍度之内。

09机制

风险如何产生与传导。

现代银行业依赖技术链条运行。一笔支付可能取决于核心银行 系统、云基础设施、网络提供商、银行卡组织和结算系统,其中许多由第三方运营。

Concentration in technology providers creates shared exposure. An outage or compromise at a major cloud, software or data provider can affect many banks simultaneously, a systemic risk that no single bank’s contract can address.

网络威胁持续存在且不断适应变化。勒索软件、数据窃取和破坏性攻击以银行及其供应商为目标。恢复取决于网络分段、可信备份和经过演练的响应。

Physical hazards affect operations too. Floods, heat and power outages can disable data centres, branches and staff access, often in the same region as borrower losses.

从防止失效转为假定失效终会发生,这一转变重塑了该学科:银行如今对服务进行端到端映射,并测试能否在设定的容忍度内恢复。

10方法与局限

证据如何收集,以及止于何处。

方法或数据显示内容其局限
服务映射各项服务背后的依赖关系随着 系统 变化,地图很快过时
情景测试Recovery within tolerances情景可能无法涵盖新型故障
事件数据中断的原因与持续时间漏报与保密
第三方登记册共同的提供方集中度对直接供应商之外的可见度有限
11方法

如何研究,以及方法无法说明的内容。

  1. 网络科学与传染银行间与交易对手传导 · 重叠投资组合与甩卖在此应用
  2. 概率与统计违约与损失参数估计 · 操作损失建模在此应用
  3. 金融与风险工程预期信用损失建模 · 交易对手风险暴露与估值调整在此应用
  4. 人工智能与机器学习具备可解释性的信贷决策 · 早期预警与交易监测在此应用
12问题

问题与解答。

什么是影响容忍度?
The maximum disruption to an important business service that a bank’s board judges tolerable, usually expressed as time, volume or customers affected. Banks must show they can remain within it under severe but plausible scenarios.
运营韧性与业务连续性有何不同?
Business continuity focuses on recovering systems and sites. Operational resilience starts from the service to customers and the market, assumes disruption will happen, and asks whether harm stays within tolerance.
为什么云服务提供商是监管关注的问题?
因为许多银行依赖少数几家服务提供商。其中一家出现故障,可能同时影响许多机构,而任何单一银行都无法通过自身合同加以管理。
Does GRA test banks’ resilience?
不。它不对任何机构进行测试。它研究共同依赖关系和方法。
Can banks share incident information through GRA?
Only in aggregated or anonymised form under the competition and confidentiality safeguards.
为什么要将物理危害纳入运营韧性?
Because data centres, branches and staff are exposed to the same floods, heat and outages as borrowers.
报告贡献

从本页贡献证据。

Reports 是联合体的季刊。您对该主题的了解可以成为下一期中的证据、案例或更正,署名归您,并在属于您本人作品时以 CC BY 4.0 发布。

Global Risks AllianceBanking NexusFinance and Insurance ReadinessScenario and Stress-Test
Banking Nexus

向联盟提出风险暴露或资本问题。

信息和咨询现已开放。每项咨询都会获得一个编号,并告知下一步;申请、付款、算力分配和注册只有在相应团队、条款和审查能力就绪时才会开放。