Guidance · BMA · 15 Sep 2025
Relevant Entities (REs) and their boards must approve annually the operational resilience lifecycle documentation, set impact tolerances using time as the mandatory metric, map resources, and identify important business services separately. Before outsourcing, REs must conduct risk evaluations and due diligence covering staff, technology, cybersecurity, data security, risk management, BCP/DRP testing and audit access; material outsourcing requires enhanced testing, monitoring, contingency plans and immediate adverse-development disclosure in written agreements. Boards must approve self-assessments and BMA statutory returns before submission.
CORPORATE SERVICE PROVIDERS, TRUST BUSINESSES, MONEY SERVICE BUSINESSES, INVESTMENT BUSINESSES, FUND ADMINISTRATION PROVIDER BUSINESSES, BANKS AND DEPOSIT COMPANIES, DIGITAL ASSET BUSINESSES, COMMERCIAL INSURERS, IIGB AND IILT INSURERS, INSURANCE MANAGERS, INSURANCE BROKERS, INSURANCE MARKETPLACE PROVIDERS AND AGENTS
September 2025
Table of Contents
1. Introduction................................................................................................................... 3
2. Board Level Governance of Op Res and Outsourcing ...................................................... 6
3. Outsourcing Materiality, Due Diligence and Risk Management........................................ 7
4. Outsourcing Agreements ................................................................................................ 9
5. Outsourcing Management Information ......................................................................... 11
6. Identifying Important Business Services........................................................................ 11
7. Mapping of Resources.................................................................................................. 12
8. Setting Impact Tolerances............................................................................................. 14
9. Communication Plans .................................................................................................. 15
10. Testing ..................................................................................................................... 17
10.1 Identifying severe but plausible disruption scenarios .................................................. 17
10.2 Test plans and testing ................................................................................................ 18
10.3 Remediation.............................................................................................................. 19
11. Lessons Learned....................................................................................................... 20
12. Self-Assessments ...................................................................................................... 21
13. Appendix A.............................................................................................................. 21
14. Appendix B – Glossary of Terms............................................................................... 22
The Guidance Note (GN) document should be read in conjunction with the Bermuda Monetary Authority (Authority or BMA) Operational Resilience and Outsourcing Code (Code).
Relevant Entities (REs) are increasingly globally interconnected when providing financial services to clients. In addition, the delivery of financial services is ‘always on’ 24 hours a day, 365 days a year. Clients expect, and have come to depend upon, real-time availability of financial services and transaction processing.
As a result, when disruptions occur, they can quickly result in harm to clients, well beyond causing temporary inconvenience. Potential harms not only include negative financial and mental health impacts on individual clients. They also include risks related to contagion to the wider financial sector and the speed at which it may occur.
The Authority has made progress towards mitigating this through the introduction of Codes of Conduct for Cyber Risk Management and through the introduction of Outsourcing Guidance. However, these are only elements of Operational Resilience (Op Res) and do not consider the wider implications (i.e. a lack thereof) for Op Res of relevant entities.
REs are already expected to have Business Continuity Plans (BCP) in place and to test these plans regularly. These plans are often ‘firm-centric’, with the Recovery Time Objective (RTO) and the Recovery Point Objective (RPO) set to suit the RE’s risk appetite rather than tailored to clients’ expectations and/or needs.
A conventional BCP often has gaps in service while recovery is underway, influenced by the RTOs and RPOs that have been set. In addition, underpinning BCP activities is a risk management exercise assessing the probability of disruptions occurring. Op Res, however, is different in that it is more holistic and client-centric.
Although the development of Op Res was driven by historical technology and cyber incidents, it is not limited in scope by either of these events. Additional events included contending with and overcoming supply chain interruptions and third-party failures, geopolitical uncertainty brought on by wars and terrorist attacks, and the global COVID-19 pandemic, including shutdowns.
Op Res is the capability of organisations to keep delivering business services that are important to their clients throughout a disruption. Its emphasis is on client-facing services rather than the interruption, or serious degradation, of an organisation’s internal services. Op Res also assumes that severe but plausible scenarios that cause disruption will occur; there is no probability component, as this is not a risk management exercise.
Op Res is frequently confused with Operational Risk. Organisations, including regulated entities, can capitalise against operational risks, but they cannot effectively capitalise against the lack of operational resilience. Operational risk focuses on identifying potential errors or failures in an organisation's operational processes and has a regulatory focus on compliance and internal controls.
Op Res is about maintaining the delivery of key services through adversity and disruptions. It also encourages organisations to have a ‘resilience by design’ mindset, where resilience becomes ingrained throughout all levels of operations rather than an afterthought. The focus of the Code is on testing and improving Op Res to cope with severe but plausible disruption scenarios. This focus allows the Authority to assess, in a repeatable, consistent manner, whether REs are sufficiently resilient.
| Scope | |
|---|---|
| Operational Resilience | BCP and IT Disaster Recovery (IT DR) |
| Op Res and Outsourcing Code | Cyber Codes of Conduct |
| • Op Res is considered an additional area that requires oversight within existing structures and/or new roles and structures, as long as governance is effective | • Governance and Enterprise-wide Risk Management (ERM) are required for BCP, IT DR and outsourcing. This is also present as a requirement in other Prudential Codes |
| Op Res • Focus is resilience only, i.e., continuing to provide an acceptable level of service without interruption and without services being degraded to a point where they can be considered unfit, despite suffering a severe but plausible disruptive event • Goes beyond IT systems/data and requires mapping the underlying People, Processes, and Facilities that underpin the delivery of the service. These five elements are collectively known as 'resources' or 'enablers' | IT DR • May mean there is an event that causes an interruption, which in turn requires recovery • Business continuity may achieve both through resilience and recovery • IT DR focus is the recovery of IT systems/data (as the acronym implies) |
| Important Client-Facing Services • Focuses on important client-facing services that can cause 'harm' if unavailable for extended periods of time • Client centric • 'Harm' is different from an inconvenience or disruption that is rectifiable in the short or medium term • The main, mandatory metric is time, which is expressed as Maximum Tolerable Period of Disruption (MTPD). REs are free to use additional metrics | Business Processes • BCP focus is on the continuity of business processes, where processes can be internal or external • Must cater to various magnitudes and severities of disruptions • Focus is on business priorities with business-appropriate metrics in terms of RTO, RPO, etc. • Business-centric priorities can be both internal and external facing, e.g., payroll |
| Resilience • The core premise is that disruptive events WILL happen to important services and therefore the RE needs to plan to be resilient accordingly. Does not require taking probability and frequency into account | Risk Management • Comprises Operational Risk Management elements with impact, probability and frequency |
| Extended Vendor Management | Vendor Management |
|---|---|
| • Mapping/understanding the five resources underpinning an outsourced Important Client Facing Service • Better understanding of sub-outsourcing to fourth/fifth parties if necessary | • Vendor management/outsourcing/third-party oversight |
| Communication Plans | Communication Plans |
| • Explicitly introduces the requirement for internal and external Communication Plans | • Communication may be part of playbooks but is not always considered |
| Testing | IT DR / BCP Testing |
| • Applies to important services • Includes the underpinning resources • Explicitly involves the vendors and vendors' resources • Applies to severe but plausible scenarios • Testing beyond (i.e., finding the limit where the RE cannot remain within MTPD) • Includes the Communication Plan | • Tests IT systems/data and business priorities • May include some elements of underpinning resources • Vendor is not always involved/considered • Scenarios vary in magnitude • Aims to recover and stay within business-specified metrics (which may or may not include time) |
| Lessons Learned | Lessons Learned |
| • Wider organisational/client-facing lessons learned | • IT and possibly business-centric lessons learned |
The Op Res lifecycle is composed of the following domains or activities:
a) Identify important business services that, if disrupted, could cause harm to clients or the wider financial sector in Bermuda;
b) Identify and document the people, processes, technology, facilities (premises) and information (data), collectively known as ‘resources’ or ‘enablers’, that support an RE’s important business services (i.e., mapping);
c) Set impact tolerances for each important business service (i.e., thresholds for maximum tolerable disruption);
d) Develop internal and external communications plans for when important business services are disrupted;
e) Test the ability to remain within impact tolerances through a range of severe but plausible disruption scenarios;
f) Remediate to ensure the RE remains within impact tolerances and conduct lessons learned exercises to invest in and prioritise activities that increase resilience and respond to disruptions effectively; and
g) Create a self-assessment document and/or submit any statutory returns the BMA requires.
For each domain, detailed guidance on the regulatory requirements, approach, expected outcomes and factors to consider is described in the sections below.
The board is responsible and accountable for the operational resilience of the RE. The board’s role in operational resilience is in line with the overall board’s function, i.e., to set risk appetite and to provide oversight and adequate challenge. The board, though not experts, should collectively have adequate knowledge of operational resilience and have access on a regular basis to the latest Management information (MI) on Op Res. The board will need to decide on and prioritise resilience initiatives, which require timely, high-quality, relevant MI.
A detailed list of what the Authority requires from the board is in the Code. The board must approve, on an annual basis, the documentation generated through the main elements of the operational resilience lifecycle, i.e.:
Identify important business services
Map resources
Set impact tolerances
Test (including testing the communication plans)
All elements of the Op Res lifecycle are covered in detail in subsequent sections of this GN. Specifically, the board should approve the self-assessment and any statutory return required by the BMA, prior to submission.
The board may appoint specific senior management roles, oversight committees and other roles as necessary to manage Op Res and may execute initiatives and activities aimed at increasing operational resilience. It is at the discretion of the board and senior management whether new senior roles and oversight committees are required or whether operational resilience is included within the scope of existing roles and oversight structures. The overriding factor is to ensure the oversight and management of Op Res is effective. It is the BMA’s expectation that the oversight, management, processes and procedures for Op Res are sufficiently detailed, adequate and proportionate to the nature of the firm’s business, size and complexity.
Note: The Operational Resilience and Outsourcing Code covers a wide variety of entities in terms of size, nature of business, and services offered. Regulatory requirements that apply to larger entities that provide a time/dependency-critical, client-facing service may not apply to smaller entities. Entities should ask themselves the following questions:
Is a sizeable client base being served?
Are the services provided time critical (i.e., would an interruption of minutes or hours cause significant ‘harm’ to clients?)
Does a client-facing service, if disrupted, pose a contagion risk to Bermuda’s financial stability?
If the answers to the questions above are ‘No’, then the entity has no service that falls under the definition of an important business service that must be operationally resilient. In these cases, unless the nature of the business changes over time, annual reviews of services, disruption scenarios, and self-assessment would be nil. For further details, please refer to:
Section 9. Identifying Operational Resilience Important Business Services within the Code
Section 6. Identifying Operational Resilience Important Business Services, bullets (a) to (f) with this GN
Notwithstanding the Op Res note above, entities are still obliged to comply with the clauses related to outsourcing and third-party oversight.
Defining what constitutes outsourcing and, consequently, material outsourcing by management in each RE will be essential for developing a credible risk-based approach to managing outsourcing risk. Not all outsourced activities will be important; therefore, management will need to determine the factors that should be used to assess which activities are important. For example, if the delivery of an activity that has been outsourced is considered ‘time critical’, then importance could be defined based on the length of time a service is not available before it damages the business’s reputation, impedes the delivery of important business services, or causes a regulatory breach or material financial loss. These could be appropriate metrics for determining the criticality in this case. Whatever metrics are used to determine materiality, they will need to be formalised and clearly articulated in the RE’s policy and procedures, with senior management able to explain the rationale for choosing them.
Where the activity being outsourced is deemed by the RE’s management not to be an important activity, the implementation of the Code provisions can be applied proportionally to the risk that a failure in the delivery of that outsourced service would pose to the RE from a financial, regulatory, and reputational perspective. The RE will need to satisfactorily explain why it is appropriate and proportionate not to fully apply those provisions.
The BMA expects the RE to undertake a risk evaluation process prior to entering an outsourcing arrangement. As part of this process, the Authority expects the RE to clearly articulate the rationale(s) for the outsourcing decisions. As such, this evaluation must set out the benefits of outsourcing and how any risks arising from the arrangement are to be mitigated/managed. There will be a specific focus on this risk evaluation by the Authority in cases where:
a) A material outsourcing is being contemplated or re-negotiated; or
b) Multiple activities are to be or have been outsourced to a single service provider, given the heightened concentration risk to which the RE would be exposed.
Subject to the Code and its related Guidance, ‘purchased services’ are services that are deemed not to be outsourced and are mainly services that other companies provide to the RE and are not part of the services and activities provided by the RE itself. Examples of purchased services include:
The supply of external advisory services to the RE that do not form part of the services and activities of the RE
Provision of external legal advice to the RE
The provision of external training for staff
The external security, management and maintenance of an RE’s premises and personnel
With regard to the purchased services definition above, if a trust company that is acting as trustee arranges the supply of investment management services to a trust, this could fall within the above definition, depending on individual circumstances. This activity would generally not be deemed to be outsourcing by a trust company acting as an individual trustee, except in cases where the trust company held itself out as providing investment management services as part of its individual trustee service. However, if the ongoing monitoring of the performance of the investment management company that is providing these services to the trust is outsourced by the trust company and acts as a trustee for another third-party provider, this would constitute outsourcing.
The BMA acknowledges that in certain instances, transferring an outsourced function back to the RE's direct remit may not be commercially or operationally feasible, and may even increase operational risk. When such circumstances arise, the RE faces significant additional risk by relying entirely on the service provider for the provision of these activities. Therefore, the Authority mandates the RE to exercise greater oversight of these arrangements in such cases and to implement appropriate contingency plans for the provision of these services by alternative providers and/or reintegrate the services back into the RE’s operations as part of its contingency planning.
Once the risk evaluation process has determined that outsourcing is the preferred option for an activity, the next stage is to conduct appropriate due diligence on the service provider.
This due diligence should include, but not be limited to, evaluating whether the service provider has the following in place:
The quantity and quality of staff with the requisite skills and experience to effectively deliver the outsourced activities, as well as the requisite legal and other authorisation(s) to perform the outsourced activity reliably and professionally throughout the life of the outsourcing
The relevant technology, cybersecurity arrangements, operational infrastructure and financial capacity to undertake the outsourcing arrangement effectively and efficiently
Appropriate information and data security to protect any and all confidential information relating to the RE and its clients
An appropriate risk management framework and controls to ensure that the carrying out of the outsourced activity is properly supervised and any risks associated with the outsourcing are effectively managed
The ability to maintain appropriate internal controls and meet regulatory requirements
An appropriate BCP and DRP and can demonstrate a successful track record of BCP and Disaster Recovery testing
The ability to provide access to all documents and data relating to the outsourced activity, its auditors and its competent authority
Additionally, the RE should consider the following:
The impact of the outsourcing arrangement on its finances, reputation and operations, or a significant business line
The risk of potential loss, temporarily or permanently, of access to important data
The degree of difficulty and time required to find an alternative service provider or bring the business activity fully ‘in-house’
In cases where the outsourcing is considered material, the Authority will require the following:
The RE’s risk evaluation assessment needs to clearly articulate the benefits and the risks of pursuing this outsourcing option
Due diligence undertaken by the RE as to the robustness and resilience of the service provider’s BCP and Disaster Recovery (DR) plans
Require more regular testing of these contingency plans at the service provider than for outsourcing where the service can be easily replaced or transferred back to the RE. This increased frequency of testing is to be explicitly included in the written agreement
More intense monitoring of the performance of the service provider by the RE than for outsourcing, where the service can be easily transferred back to the RE. This increased monitoring is to be explicitly included in the written agreement
Immediate disclosure, as part of the written agreement, to the RE by the service provider at the point when it first becomes aware of any legal, operational, technological, financial, resource or regulatory adverse development that may affect the service provider’s ability to provide the outsourced activity
Require the development by the RE of more detailed contingency plans that could be utilised if the service provider is unable to provide the outsourced activity. In this case, the contingency plan would relate to the ability to transfer the activity to other service providers in the same jurisdiction in a timely manner
At a minimum, the policies and procedures should include the following:
An effective method for identifying and quantifying concentration risk exposures related to outsourcing activities, including dependencies on specific outsourcing service providers, geographic locations and important activities
Regular risk assessments evaluating the impact of concentration risk on the RE’s operations, financial condition and operational resilience
Robust and enhanced monitoring frameworks for identified outsourcing concentration risks
Depending on the activity being outsourced, the Authority expects the written agreement to specify the following:
The activities to be outsourced and the laws/regulations applicable to the agreement
The responsibilities of the RE and service provider in the agreement
Relevant policies, procedures and controls to ensure the ongoing security and confidentiality of information provided by the RE to the service provider
An obligation imposed on the service provider to comply with all relevant data protection, and data privacy rules and regulations
An obligation imposed on the service provider to maintain appropriate risk management standards and internal controls throughout the life of the outsourcing
An obligation imposed on the service provider to provide regular updates on its financial soundness, and confirmation that it retains the human expertise, and technological and operational capacity to provide the contracted activities throughout the life of the outsourcing
An obligation to impose a material adverse change disclosure on the service provider to immediately disclose to the RE any legal, operational, technological, financial, resource or regulatory adverse development that may affect the service provider’s ability to provide the outsourced activity on an ongoing basis
The agreed upon quantitative and qualitative service level standards and performance targets to be met by the service provider in performing the activities, together with the method and frequency by which these quality standards and performance metrics will be monitored by the RE throughout the life of the outsourcing
Specify the reporting and escalation process where performance standards are not met, and the dispute escalation and resolution process agreed upon by both parties
An obligation on the service provider to provide the RE with regular updates on the following:
a) The adequacy of its BCP and DRP;
b) Any material changes in its BCP or DRP that would affect the provision of the RE’s activity; and
c) The results of the regular testing of its BCP and DRP (in conjunction with the RE, if requested)
Whether or not sub-contracting is allowed in the agreement, and the conditions and liabilities imposed on the service provider and the sub-contractor, where this is allowed
The conditions, triggers or thresholds that would allow either party to terminate or exit early from the agreement
An obligation imposed on the service provider to provide access to all documents and data relating to the outsourced activity to the RE, its auditors and its competent authority, as well as providing access to the business premises of the outsourcing service provider for these parties if required
In the case of both intra-group material and non-material outsourcing, any written outsourcing agreement between the RE and the group service provider may be supplemented with other group documents (including but not limited to group policy and procedure documents, Performance Level Agreements (PLA) and/or Service Level Agreements (SLA). This is provided the RE can demonstrate to the Authority, if requested to do so, that the provisions contained in the supplemental group documents are sufficiently robust and can be relied on by the RE to deliver the relevant protection/action, as needed.
Management information should, at a minimum, contain the following:
A summary of all outsourcing arrangements, detailing the scope, nature and importance of outsourced activities
An evaluation of the risks associated with each material outsourcing arrangement, including legal, compliance, reputational, operational and resilience risks
An overview of the risk mitigation strategies implemented for each material outsourcing arrangement
An update on the material outsourcing arrangements that are aligned with a board-approved outsourcing risk appetite
Notification of any material changes to the contractual agreements between the RE and the outsourcing service provider for material outsourcing arrangements
Review of BCPs and DRPs for material outsourcing arrangements
A summary of the results of the most recent BCP and/or DRP testing, including an overview of remediation efforts in response to any deficiencies identified during the testing
An analysis of the service level performance metrics as agreed between the RE and the outsourcing service provider
Reporting on any incidents, breaches or disruptions of the material outsourcing arrangements, including any related operational losses and actions to be taken to resolve and prevent future incidents
An assessment of compliance with regulatory requirements and the reporting of any breaches of these requirements
Identification of emerging risks and trends in relation to outsourcing, including technological advances, regulatory changes and market developments
Business services deliver a specific outcome or service to an identifiable user external to the firm and should be distinguished from business lines, which are a collection of services and activities (e.g., a specific product offering).
Banks must identify important business services that may pose a risk to their clients. Insurance companies must also identify important business services that may pose a risk to policyholders’.
Additional factors not listed in the Code that the RE should consider when identifying its important business services may include:
a) The sensitivity level of the data held;
b) The potential impact on the RE’s financial viability if any disruptions to clients occur;
c) The potential impact of reputational damage to the RE and reputational damage that extends to clients or, in the case of insurance, policyholders;
d) The potential impact on the orderly operation of the wider financial sector;
e) Whether or not disruption to the services could amount to a breach of a legal or regulatory obligation; and
f) Clients that are important to the financial stability of Bermuda.
An RE’s important business services will be a relatively short list of external-facing services for which the firm has chosen to build high levels of operational resilience in anticipation of operational disruption.
An RE should not identify a collection of services as a single important business service when identifying its important business services. For example, ‘self-service banking’ comprises telephone banking, internet banking, mobile apps, and ATMs.
REs should identify each distinct relevant service separately and ensure due process is followed for Code compliance. Continuing with the ‘self-service’ example above, four distinct services are identified. These may share or have in common underlying ‘resources’ in terms of people, processes, technology systems (IT systems), facilities (premises), and information (data).
In the case of insurance services, the insurance company should consider providing an appropriate degree of policyholder protection. This protection should mainly concern the impact that a disruption in service may have on policyholders, which in turn depends on the type of product, type of policyholder, and the nature of any adverse effects on the policyholder and the wider market if policies are not honoured.
REs are required to identify and document the necessary people, processes, technology systems (IT systems), facilities (premises), and information (data), collectively known and referred to as the ‘resources’ (enablers) required to deliver each of their important business services. The terms in brackets are used interchangeably with their counterparts within the Code and this GN.
Below is a broader but not definitive description of the resources. However, REs are encouraged to include any wider definition of the resources in their mapping if it better serves the mapping exercise and the quality of the information as the outcome.
People – Under this category, people within the organisation and its vendors support and enable the provision of important business services. These may include key staff and contractors with specific technical skills/knowledge/experience, decision makers and approvers, overall senior management accountable for oversight, and staff responsible for implementing and monitoring controls. Substitutability, knowledge sharing and transfer, teams operating in silos, teams’ strength and size, staff churn, attrition and hiring lead times, skills scarcity and competitor demand for similar skills in the market. Niches, specialised skills and roles that come into play during an incident (e.g., cyber forensic analysis) also should be considered. Please refer to the ‘Testing’ section and succession planning within this GN.
Processes – A process is a structured set of activities designed to produce a specific output. Organisations should have end-to-end visibility of the chain of processes and activities leading up to and enabling the delivery of important business services. Internal and external processes, as well as internal and external inputs and outputs, should be clearly defined and understood. Processes may be technology-driven or otherwise.
Technology Systems – Information Technology (IT) systems and the underlying architecture that supports the provision of the service. Similar to the processes above, several IT systems may be involved that are interconnected by the flow of data and enabled by the underlying architecture of networks, servers and so forth. Due consideration should be given to the added complexity when IT systems and underlying architecture are provisioned wholly or in part by third parties.
Facilities – Office locations, data centres, printing and mailing centres, credit card production/statements, client communications/helpdesk, Network Operations Centre (NOC), Cash Depot/Vaults.
Information – Any data, feeds or material that is required by an organisation to deliver a service. These may include stock exchange, foreign exchange and other market information feeds. Consideration should be given to whether or not the information is stored in different formats or undergoes several iterations of data manipulation and/or aggregation prior to its use for delivering a service. Refer to ‘Processes’ above and related activities.
The level of detail required in the documentation should be proportionate to the size, scale, and nature of the business services offered by the RE. The BMA will not prescribe a rigid approach or methodology. The RE must ensure that the mapping exercise is usable and beneficial to itself.
Intra-group arrangements and third parties increase the complexity of the mapping exercise. These need to be assessed and mapped accordingly. The Code itself provides an extensive explanation of how the firm must comply with regulations irrespective of outsourcing arrangements. The BMA published Outsourcing Guidance and Cyber Codes, which outline regulations in this domain.
Intra-group arrangements, sharing group resources and provision by third parties provide benefits to REs, i.e., efficiency, cost savings, expertise, etc. The mapping complexity arises if there is less visibility of how the service provider operates internally, its resources and its own resilience. For intra-group arrangements, there may be more visibility in this regard. However, both intra-group and third parties may have conflicting and competing priorities during a disruptive incident when compared to BAU. If a service provider is providing a service to several firms and that service is severely limited during a crisis, particular firms may suffer as a result.
To add a further layer of complexity, third parties may outsource to fourth parties. Though not mandated in the Code, REs should have a clear understanding of how fourth/fifth parties’ capabilities may affect the third party’s ability to provide an important business service.
An RE is still responsible for ensuring it remains within impact tolerance if a third-party provider fails or causes the firm to be outside the impact tolerance. However, it is recognised that Bermuda has specific infrastructure limitations, e.g., having one power station and other Critical National Infrastructure (CNI) restrictions. It is acknowledged that there will be instances where firms will be out of tolerance despite all the resilience put in place, due to circumstances and resources beyond their control.
The internal services that support an important external business service are to be considered and mapped under the ‘Process’ resource. The chain of internal services performed must be resilient; otherwise, an internal service failure would break the chain and put the important business service out of tolerance threshold. Testing and improving resilience measures will be required for any resources that put the important business service at risk.
The mapping should be reviewed on an annual basis or when there is a material change. Due to the varying size, nature and complexity of the REs, the BMA intentionally does not prescribe what qualifies as a ‘material change’. REs are expected to have their own definition of a ‘material change’. In principle, mergers and acquisitions, new business lines, changes in key personnel, changes in key vendors or intra-group arrangements, tighter SLAs by counterparties, and notable premises or facilities changes (e.g., migrating a data centre) should be considered by firms as triggers for a mapping exercise. The exercise should not be done from scratch; instead it should factor in recent material changes that have been evaluated.
The expected outcome of the mapping exercise is for the RE to have a clear understanding of all the resources required for each important business service in order to:
a) Conduct testing (i.e., identify and remedy vulnerabilities and shortcomings); and
b) Identify key personnel and develop short-term and long-term plans (i.e. succession planning) should those key personnel be unavailable.
Mapping exercises are expected to be developed and refined after each iteration.
Impact tolerance is the maximum level of disruption to an important business service as measured by the length of time, in addition to other metrics. Since the main metric is time, MTPD is used to measure this.
As per previous sections of this GN, Op Res is not a risk management exercise. Therefore, neither the cause nor the probability of a disruptive event needs to be defined; instead, it is assumed that the disruptive event will occur.
It is not mandatory to consider the frequency at which disruptive events are likely to occur. Impact tolerance is set for a single disruption. REs may, at their discretion, consider multiple disruptions or frequent disruption scenarios if they help inform the impact tolerance setting exercise. However, it is not the aim of Op Res regulations to introduce unwarranted complexity.
REs must set an impact tolerance for each of their important business services. Again, time should be used as the main and mandatory metric. Additional metrics should be considered for inclusion.
Examples include transaction volume, total transaction value, settlement deadlines, and legal and/or regulatory obligations.
Additional factors not listed in the Code that an RE should consider when setting impact tolerances may include the following:
a) The sensitivity of data held and any potential loss of confidentiality, integrity or availability of data;
b) Clients’ resulting financial losses;
c) The RE’s financial loss and the potential impact on its financial viability in the event of further disruptions to clients;
d) The potential to cause reputational damage to the RE in terms of that reputational damage extending to clients, including in the case of insurance policyholders;
e) The potential impact on the orderly operation of financial markets, including contagion risk;
f) Client confidence and wider market confidence; and
g) Whether or not the disruption to the services could amount to a breach of legal or regulatory obligations
As the impacts of a disruption should be clear, the clients (users) of the service should be identifiable. There should be clear demographics or characteristics to identify affected client groups.
Due consideration must be given to the possibility of a common shared resource (e.g., data centre) being disrupted, which, in turn, causes several important business services to become out of tolerance. This may be particularly relevant where services are shared across organisations that are part of the same group of companies. This is another reason why the mapping of resources must be comprehensive and have an adequate level of detail in order to provide the RE’s management with the information they need to take Op Res steps.
REs should also examine the complexity of the services they provide and factor in other variables, such as peak times or peak periods (e.g., Christmas holidays), any periods of time when the firm’s operations and viability are most in jeopardy and the period of time when service unavailability poses contagion risks that may impact Bermuda's financial stability.
For REs that are insurers, policyholder protection and any adverse effects on the policyholder and the wider market if policies are not honoured should be considered.
The Code states that REs must ensure that staying within chosen impact tolerance during disruption, does not cause harm in other ways. This clause considers the repercussions and unintended consequences that may happen when the primary aim is to remain within impact tolerance per se, rather than focusing on avoiding harm to clients. In certain situations, such as corrupt or incorrect data, cyber-attacks and other scenarios, the risks of keeping a service running are higher than suspending/shutting down the service, which can result in breaching the impact tolerance threshold. A sensible approach must be considered and, when in doubt, discussed with the BMA.
Communication plans have always been an essential component of BCP and IT DR. They are even more crucial for Op Res as its scope is more extensive in terms of planning, scenarios and the number of stakeholders involved. Timely and effective communication during a disruption or crisis is essential for:
a) Co-ordinating internal resilience efforts, including escalation paths and briefing senior management;
b) Co-ordinating with vendors, suppliers, utility providers and emergency services;
c) Keeping the clients/end users up to date with the status and progress of the disruption;
d) Directing clients to alternative services and channels where substitutability is an option;
e) Briefing and coordinating with the regulator;
f) Co-ordinating internally to ensure the RE’s staff are not independently issuing publicfacing statements or updates unless vetted and approved as part of the planned communication channels; and
g) Managing the media, which may include:
o Issuing press releases and/or conducting interviews through various media channels, as appropriate;
o Monitoring the media’s narrative in the reporting on the disruption;
o Monitoring social media and providing responses, as appropriate.
An RE must have a communication strategy in place to respond to, manage and mitigate disruptions. The communication strategy must include internal and external communication plans for all the severe but plausible scenarios that the firm envisages for its important business services. These may include pre-planned statements, also known as ‘canned responses’ tailored for various delivery channels such as tweets, web splash pages, etc.
The communication plans must provide clear, timely and relevant information (cause, extent, impact) to stakeholders. Stakeholders include but are not limited to clients, third parties and vendors, intra-group, counterparties, shareholders, regulators and emergency services. A communication plan set at the group/overseas level needs to be reviewed and tailored to be practical and applicable at the local level.
REs’ communication plans must include:
a) Escalation paths and timeframes/other triggers for escalation; and
b) Decision makers, key individuals and main contacts at suppliers, counterparties, regulators, et al.
Communication is more effective when it includes the time or time interval between updates (e.g., every two hours on the hour). The communication update schedule should be strictly adhered to. Even when there are no changes to report, the updates should be released stating this. Regular and clear updates engender trust amongst stakeholders and provide the narrative from the RE’s perspective.
Where appropriate, external communication must include warnings and/or updates if only indirect communication is possible (e.g., website splash page).
Communication plans must be tested as part of scenario testing. Key vendors must be included in the communication plan tests.
As described in the Code, the next stage after setting impact tolerances for the important business services, is to test the ability to remain within tolerance when severe but plausible disruption scenarios occur.
It is assumed that disruptions will occur. Consequently, as previously discussed, Op Res is not a risk management exercise in assessing probability. Therefore, the focus of testing should not be incident prevention but resilience. REs need to focus on how they will continue to provide important business services at an acceptable level (i.e., with services not degraded to an extent to be considered unusable or interrupted) despite suffering a severe but plausible disruption scenario.
Identifying the scenarios to be considered for testing is an important exercise in and of itself. REs should consider the following:
a) The nature of the business;
b) The local environment they operate in, in terms of vendors (limitations of), counterparties, natural phenomena such as hurricanes, telecoms and utilities such as electricity;
c) Any previous incidents within the RE, the wider group, if applicable, the financial sector within its home jurisdiction, and other jurisdictions where it operates;
d) Group considerations, if applicable, such as shared infrastructure and other dependencies, group incident support, group priorities and operational, functional, and technical service performance levels and requirements, including relevant key performance indicators (KPIs);
e) Upstream and downstream obligations (e.g., settlement, policy handling for insurers);
f) Regulatory and legal requirements and compliance;
g) Reputational risk.
Possible disruptive scenarios that REs should consider, among others, are:
a) Corruption, deletion or manipulation of data critical to the delivery of its important business services (also refer to the Cyber Code of Conduct);
b) Unavailability of facilities, premises or key people;
c) Unavailability of third-party services or vendors, which are critical to the delivery of its important business services;
d) Disruption to, or by, other market participants or market contagion;
e) Loss or reduced provision of technology or the IT services that underpin the delivery of important business services;
f) Unique points of failure (e.g., utility supplies, Critical National Infrastructure);
g) Dependency on Group and vice-versa, as applicable;
h) Catastrophic natural events such as hurricanes, earthquakes and pandemics; and
i) Political disruption such as civil unrest, riots and war.
Test plans provide a documented, repeatable way of gaining reassurance on the level of resilience for important business services. Initial tests are used to establish a baseline. It is expected that tests will evolve in detail, complexity and realism after each iteration, as lessons learned are fed back into the test design. Any test assumptions should be realistic and not a shortcut to test success, i.e., to stay within impact tolerance, as the scope of operational resilience is severe but plausible scenarios.
In addition to the test plan design considerations listed in the Code REs should also consider:
a) The size and complexity of the RE itself and the wider group, if applicable. A proportional approach is expected. Some REs will identify more services and more disruptive scenarios than others, which would result in undertaking considerably more tests;
b) Substitutability of services and resources is a requirement in the Code itself. This should be noted by REs subject to the Code;
c) The frequency with which changes are made to operations or particular services and associated underlying resources. These changes should drive the frequency of testing and
d) How often does the operating environment change in terms of the financial landscape, vendors, utilities and the natural environment. Environmental changes, in turn, could also give rise to different threats or disruptive scenarios and, therefore, should also be considered.
Adequate oversight of third parties and the involvement of third parties in testing are key. The Code outlines People, Processes, Technology Systems (IT Systems), Information (Data) and Facilities (Premises) as the five resources (or enablers) underpinning an important business service. Any of these resources may be outsourced by an RE to third parties, including other members of a group of companies. The Cyber Code of Conduct and Outsourcing Guidance mainly set out Technology Systems and Data Outsourcing Regulations. The Operational Resilience and Outsourcing Code extends and sets out vendor oversight expectations for People, Processes, and Facilities.
REs should also consider vendors’ own impact tolerances, as well as fourth and fifth-line vendors (i.e., the vendor’s vendors and the vendor to the vendor’s vendor), especially in cases of commonality where contagion is possible (e.g., vendors within the same building/premises, vendors using the same products or hosted at the same cloud provider).
REs and their vendors should take reasonable steps to support each other in order to achieve resilience goals. REs should endeavour to secure the availability of vendors for testing.
REs may consider scenarios where a vendor is unavailable for a considerable period of time or possibly the total failure of a third party and a forced termination of the service/dependency. REs should also assess whether international events may disrupt the local vendor services (e.g., when a vendor is headquartered or runs substantive business elsewhere, especially in high-risk countries subject to natural disasters or political instability).
Similar to other industries (e.g., composite materials), testing should be repeated with increasing severity past failure. Tests can be tweaked to prolong the disruption period, reduce resource availability further or introduce other variables. REs should know the thresholds beyond which impact tolerances cannot be met.
It is not a regulatory expectation that all scenarios are successfully contained all the time. This is especially the case for scenarios over which the RE has no control, such as CNI.
Testing must not risk jeopardising live systems and services. If testing involves live systems or is the appropriate way to realistically test, an RE should assess the risks prior to deciding whether to proceed with the planned test or consider an alternative method of testing. This risk assessment should be retained as part of the testing iteration.
Testing outcomes should feed into several areas that include the following:
a) An increased focus on recovery and response arrangements. This, in turn, may be subdivided into improving technical controls, bolstering underlying resources (refer to section 7. “Mapping of Resources”) and improving communication (refer to section 10. “Communication Plans”);
b) Management Information (MI) to the Board and Senior management to aid prioritisation and investment decisions; and
c) Updating and disseminating lessons learned.
The Code sets out the remediation regulatory expectations. REs must take a responsible and sensible approach to remediate any issues that would cause important services to exceed impact tolerances in a real-life, severe, disruptive event.
Responsible and sensible approaches include but are not limited to:
a) Ensuring tests are realistic and as accurate as possible;
b) Taking test results seriously. This goes beyond conducting a ‘box-ticking exercise’ to comply with regulations;
c) Acting in a timely manner to rectify failures and shortcomings. Remedial action should also be time-bound;
d) Investing in resilience. Op Res should be part of the organisational culture and be cascaded down from the top;
e) Prioritising those services that are furthest outside the impact tolerance, where possible, as these are deemed to be the highest-risk services;
f) Taking into consideration possible repercussions for the wider financial sector, contagion and policyholders (in the case of insurers); and
g) Ensure that remedial action is fit for purpose and caters to complex services, group arrangements and overseas provision of services.
Embedding operational resilience within an RE and the services it provides is not a one-off exercise. Op Res comprises a set of dynamic activities, including assessment, testing, and integrating the lessons learned back into the RE’s operations as part of a continuous cycle of improvement. This should be part of the firm’s overall culture and approach.
REs should have clear processes, procedures and management information on their important business services, impact tolerances, possible severe but plausible disruptive scenarios and the ability to remain within tolerance after the first iteration of implementation. The expectation is that REs continue to refine, add to, and fine-tune every aspect of Op Res year-on-year in order to strengthen their own resilience and that of the wider financial sector within Bermuda.
The most valuable lessons can be learned not from simulated testing but by evaluating and reviewing the performance and behaviour of the organisation and its services in the aftermath of real-world disruptive events. These provide the opportunity for an RE to assess whether or not the service remained within tolerance, whether the testing was credible and accurate and whether the scope covered all the variables and resources necessary. Overall, regular Op Res testing is an opportunity to recalibrate and re-run prior tests, improve resilience where it matters most and assess what can be further improved upon. This assessment should include, for example, questions such as these below:
a) Was the disruptive scenario that occurred previously identified as a possibility, or was it not even featured in the testing?
b) Were key people available?
c) Did the disruption highlight any gaps among BAU operations and key personnel in the event of an emergency? Were resources stretched?
d) Would the outcome be different or worse if the disruption had lasted longer?
e) Were there any chain reaction effects that occurred in real life, but were not considered during testing?
f) Were new bottlenecks in resources identified? For example, did the main means of internal communication that is vital to coordinating the response become unavailable too?
g) Did third parties, including vendors, utility companies and emergency services, if applicable, behave and react as expected?
h) Was the reaction and mitigation timely?
REs should undertake their own post-mortem self-assessment, using questions applicable to their specific business, size, complexity, and services offered to determine the lessons learned.
To demonstrate compliance with the Code, REs are required to document their resilience-related activities and outcomes, prepare a self-assessment annually, and make the self-assessment available to the BMA upon request.
Self-assessments should demonstrate the board's confidence that the RE is able to provide important business services within the RE’s impact tolerance year over year. The board is held accountable for Op Res and is responsible for approving the self-assessment.
The expected contents of the self-assessment are listed in the Code. The creation of a self-assessment and making it available to the Authority upon request are regulatory requirements. However, the self-assessment should also serve as one of the sources of Op Res MI for the board and senior management, and its level of detail should reflect the complexity, size and nature of the services provided. The self-assessment ties the components of the Op Res lifecycle together and helps rationalise how services are identified, resources mapped, and vulnerabilities captured. Most importantly, the self-assessment should document the efforts and investment the RE is making to improve resilience.
For important business services that are outside impact tolerance thresholds, the self-assessment must describe the actions the RE is planning/undertaking to bring the services within the impact tolerance and the timelines for completion.
The following are provided as guidance to assist REs with implementing the Code.
Important activity - An operational function that is considered important if a flaw or failure in its performance would significantly affect an RE’s ongoing compliance with the terms and obligations of its license or registration, other duties under its associated Act, and, or clients and dependent third parties. Such a defect could also have a material impact on the financial performance, stability, or continuity of its operations and activities.
Important Business Service - A service that an RE provides to external clients to deliver a specific outcome. Important business services are defined as follows if they are disrupted:
They can cause significant ‘harm’ to clients, in particular depositors and policyholders
They can cause significant ‘harm’ to other stakeholders
They pose a contagion risk to Bermuda’s financial stability
Material outsourcing - An outsourcing arrangement where an important activity, as determined by senior management of the RE, has been outsourced to a third party.
Outsourcing - An arrangement in which the RE uses a third-party, i.e., (the outsourcing service provider), to perform activities on an ongoing basis that are integral to the provision of services by the RE itself that could otherwise be undertaken by that RE.
Outsourcing agreement - A written, legally enforceable agreement that sets out contractual terms and conditions governing relationships, obligations, responsibilities, rights and expectations of the contracting parties in an outsourcing arrangement. This term also covers operational, functional and technical service performance levels and requirements, including relevant key performance indicators ( KPIs) between the contracting parties..
Outsourcing service provider - A third-party that provides a service to an RE. This third-party entity may or may not be licensed and may be either an affiliated entity within the RE’s own corporate group or an entity external to the RE’s group.
Sub-contracting/sub-outsourcing/chain outsourcing - An arrangement where the outsourcing service provider that has an outsourcing arrangement with an RE to perform an activity, then subcontracts the provision of all or part of that activity to other service providers.
BCP – Business Continuity Planning
BAU - Business as Usual
DRP – Disaster Recovery Planning
CNI – Critical National Infrastructure
Code – Operational Resilience and Outsourcing Code
MI – Management Information
MTPD – Maximum Tolerable Period of Disruption
PLA – Performance Level Agreement
RE – Relevant Entity – an entity licenced by the Authority under relevant acts
RTO – Recovery Time Objective
RPO – Recovery Point Objective
SLA – Service Level Agreement
Note from RegAlert. AI assistants can read this document in full, and search 70,000+ more, through the RegAlert MCP connector (https://mcp.regalert.today/mcp). Free with an account. How to connect ChatGPT, Claude or Cursor.
More like this from BMA
BMA published 5 documents in the last 30 days. We email you each new one the day it's published.
Source: Bermuda Monetary Authority — original document · Summary generated with machine assistance and checked by an AI reviewer before publication; the authoritative text is the regulator's original document. How RegAlert works
In RegAlert: added · updated