Given the range of outsourcing arrangements which may fall under the updated guidelines, the need for national competent authorities to amend supervisory practices within existing rules, and the introduction of new requirements and resulting implementation of new processes / information gathering exercises required for compliance, the proposed application date of 30 June 2019 in paragraph 12 and transitional deadline of 31 December 2020 in paragraph 13 for the documentation of all existing arrangements would be extremely challenging to achieve.
As proposed, substantial time and considerable resources will be needed for outsourcing institutions to fully comply. This work may include potential changes to governance and internal policies, adjustments to or new builds of an institution’s register, varying development lifecycles to update IT systems around potentially complex control systems and hold notices for any outsourcing projects in flight to verify compliance with the final guidelines. Importantly, a significant volume of work is expected related to the review of all existing outsourcing arrangements, and the potential need to update and negotiate contracts with service providers. These issues are further complicated by the fact that the proposed deadlines do not apply a risk-based or proportionate approach, as they require adherence to all provisions by the respective dates regardless of the critical or non-critical nature of the outsourcing.
It is our understanding that the final guidelines are expected to be published officially by the end of February 2019, leaving institutions with only four months before the guidelines apply. In order to provide sufficient time to implement the required changes, and to allow time for institutions to communicate with national competent authorities on any local considerations, we recommend the EBA revise the application date in paragraph 12 to 31 December 2019 at earliest.
The EBA should also consider a phased approach to compliance for existing arrangements, based on the potential risk impact of the outsourcing. Completion of documentation of critical and important outsourcing should be prioritised, while any existing, non-critical outsourcing service should not be considered as in scope for mandatory updates unless such contracts are subject to renewals or relevant amendments (typically undertaken every 2-5 years).
We therefore recommend that the transitional deadline of 31 December 2020 only apply to existing outsourcing arrangements deemed critical or important. Non-critical outsourcing arrangements would continue to be addressed according to their typical review and update cycle.
The alignment in section 2 of Title I of the guidelines, relating to outsourcing within group application with the EU Capital Requirements Directive / Regulation (CRD/R) is helpful, however, the proposed language infers that an outsourcing institution would be required to apply the EBA requirements in full to all entities, including third country entities within a group, unless foreign law expressly prohibits such measures. In order to avoid confusion over potential extra-territorial reach, clarification is needed to confirm that the application of the guidelines are intended for authorised entities located in the EU. We therefore recommend that paragraphs 17 and 18 be amended as follows:
'17. Institutions and payment institutions located in the EU which are subsidiaries of an EU parent undertaking or of a parent undertaking in a Member State to whom no waivers have been granted on the basis of Articles 7 and 10 of Regulation (EU) No 575/2013 or of Article 21 of Directive 2013/36/EU should ensure that they comply with these Guidelines on an individual basis in accordance with Article 109 (1) of that Directive.'
'18. In accordance with Article 109(2) of Directive2013/36/EU, these Guidelines should apply on the sub-consolidated and consolidated basis. For this purpose, the EU parent undertakings and the parent undertaking in a Member State should ensure that internal governance arrangements, processes and mechanisms in their subsidiaries located in the EU, including payment institutions are consistent, well-integrated, and adequate for the effective application of these Guidelines at all relevant levels.'
We welcome the clarity provided in paragraph 19, in conjunction with paragraphs 46 – 47 in section 8, that institutions within a group may rely on a centrally-maintained register as long as each institution can easily draw from the central register all relevant information required under the guidelines.
As drafted, however, paragraph 19(b) could be interpreted to imply that: i) each competent authority is expected to have ‘their’ own register, maintained at the institution; and ii) the EBA intends to regulate the relationship between non-EU subsidiaries and their non-EU supervisory authorities.
We believe that neither was the intention of the EBA as there is no need for such measures if there is a centrally maintained register from which timely extracts can be supplied. Therefore, to avoid potential confusion in the implementation of the final guidelines, we recommend that paragraph 19(b) be amended as follows:
‘where the register of all existing outsourcing arrangements as referred to in Section 8, is established and maintained centrally within a group, the competent authorities, all institutions and payment institutions should be able to obtain an extract from that register providing a view of those outsourcing arrangements (extra and intragroup) relevant to entities within their jurisdiction without undue delay.’
The EBA also clarifies that intragroup outsourcing should be subject to the same regulatory framework as outsourcing to service providers outside of the group. The EBA acknowledges that intragroup outsourcing may allow for ‘a higher level of control over the outsourced function which they could take into account in their risk assessment’. However, the requirements laid out in the guidelines imply that intragroup arrangements can be equal or even riskier than external outsourcing, and in particular highlights concerns over potential conflicts of interest. These conflicts are not further specified or explained – especially in comparison to conflicts which may exist with a third party vendor – in the draft guidelines
While we support the application of the guidelines to intragroup arrangements, we are concerned that insufficient recognition is given to the benefits afforded by integrated risk management processes for intragroup arrangements nor the enhanced control framework available to service recipients. In comparison to outsourcing to third parties, these internal processes and controls result in a lower risk profile for intragroup outsourcing.
For example, in line with Articles 109(2) and 74 of the EU Directive 2013/36, each subsidiary within the group is required to operate in-line with globally applicable group policies and standards on customer protection, risk management and internal controls. Furthermore, the internal audit system is applied to all subsidiaries worldwide, with intragroup service documentation usually based on standardised contract templates centrally issued by the legal department. These benefits are present regardless of the location of the service provided, so third country intragroup providers should not automatically be assumed to be riskier than providers located within the EU.
In this regard, a differentiated approach should be applied which provides preferential treatment to intragroup outsourcing, in order to more accurately reflect the actual risk posed by these arrangements. The final guidelines should provide a more proportionate, risk-based approach which differentiates between third party and intragroup outsourcing, with regards to, but not limited to: the criteria for assessing criticality or importance; due diligence; preconditions for outsourcing to third country entities; and exit strategies.
We welcome the examples provided in paragraph 23 of activities which should not be considered outsourcing; however, it would be helpful to provide additional clarity by noting that the list in paragraph 23 is non-exhaustive. This should be complemented with a more detailed list of examples in the annex of the final guidelines, such as the list presented in the European Banking Federation’s (EBF) response letter to the EBA.
As outsourcing is likely to play an increasingly important role in the financial system as new business models, partnerships and technologies gain traction, it would benefit both industry and competent authorities to ensure that only the services which pose risk to the outsourcing institution are subject to the updated guidelines. The EBA should therefore consider reviewing this list in cooperation with industry participants every two years and update as needed.
In regards to the requirements outlined in paragraphs 25 – 26 on the outsourcing of activities that require authorisation or registration by a competent authority in the Member State the institution is authorised, the current wording risks an overly restrictive approach which could constrain the outsourcing of specific elements of processes or services for efficiency and operational reasons.
It would be useful to clarify that the restrictions and requirements in these paragraphs do not affect outsourcing of parts of banking activities which are subject to authorisation, but are instead limited to licensed activities themselves. For example, it is not clear why back office functions and processing / operational functions should not be outsourced to an entity that is not licensed for the entire activity or is located in a third country, as the licensed activity still remains under control of the institution. This is especially relevant in the context of payment or securities settlement activities, where outsourcing to intragroup or external ancillary service providers is very common. An institution may also be required to retain a service provider in a third country for cross-border business to partially support certain elements of the business, e.g. for KYC due diligence. We therefore recommend that paragraphs 25 and 26 be amended as follows:
'Without prejudice to the requirements within Title III, institutions and payment institutions should ensure that, if banking activities or payment services that require authorisation or registration by a competent authority in the Member State where they are authorised are to be outsourced in full or to a material extent, that they are only outsourced to a service provider located in...'
Furthermore, we are concerned that the criteria set out in paragraph 26 for outsourcing arrangements in third countries will unnecessarily restrict the range of providers available to institutions. While we agree on the importance of the home supervisor’s ability to supervise the outsourcing institution, the requirements in paragraph 26(b) applies a singular solution through the use of a formal memorandum of understanding (MoU).
Recognising that this requirement should only apply to the outsourcing of authorised banking activities or payment services in full or to a material extent (as per the above recommendation) and thus narrow the scope of impacted arrangements, limiting the supervisory mechanism to MoUs would nonetheless be problematic. This could restrict a range of countries where cooperation agreements are not yet available, where the timing of their finalisation is not known or where existing cooperation agreements may be revoked.
We believe a more suitable approach would be an outcomes-based one, as outlined on page 57 of the guidelines as a potential policy option, which relies on an institution’s assessment that outsourcing to service providers in a third country would not prevent or undermine an EU national competent authority’s ability to effectively supervise the outsourcing institution and their outsourcing arrangements. This would provide greater flexibility and certainty to institutions on the continued viability of service providers located in third countries, while maintaining effective supervisory capabilities.
If the guidelines however maintain the existence of an MoU as a precondition, the establishment and maintenance of a public register for MoUs by the EBA would also provide much needed transparency on cooperation agreements which meet the requisite requirements in paragraph 26. This would include updating and providing clarity to outsourcing institutions on scenarios where an MoU is withdrawn or amended.
Additionally, we recommend the EBA also provide relief from the requirements set out in paragraph 26 in the case of outsourcing to a service provider in third countries, when such provider wholly belongs to the same group as the outsourcing institution. These entities would therefore be part of the organisational structure and subject to the risk management policies of the group. Intragroup contracts as well as group governance will ensure full compliance with the regulatory requirements of the service recipient. This could be addressed with the addition of a new paragraph 26(d):
'outsourcing to a service provider located in a third country that is wholly owned by an institution or payment institution and subject to the risk management policies and governance processes of the group would however be exempt from the requirements set out in paragraph 26(a), (b) and (c).'
With regard to the governance arrangements set out in paragraph 32(g), in combination with requirements set out in paragraphs 89-91, such sections would expressly require for an alternative service provider or the ability to reintegrate an outsourced critical or important function, and to ensure such exit plans are comprehensive, documented and sufficiently tested.
While we generally agree and support such requirements, it is unclear how the proposed testing requirements would work in the case of complex services like a data centre outsourcing, where migration activities would require significant and complex planning efforts and typically take multiple years to complete. Exit plans are based on assumptions and ‘moving targets’, and are therefore hypothetical, i.e. cannot be ‘tested’ without actually a full effort of, for example, going to market with a second provider. Therefore, to better reflect such scenarios we propose to amend the ‘testing’ language in paragraph 90(a) with a requirement for comprehensive and plausible documentation as follows:
'develop and implement exit plans that are comprehensive and documented in a plausible manner'
It is also worth noting that in certain circumstances the identification of a suitable alternative service provider is not always possible. As highlighted in the EBF’s response letter, this may include the use of SWIFT and may in the future include certain innovative technologies which require specialised expertise. In such situations we believe the requirement in 32(g) should not immediately preclude the outsourcing arrangement, as long as a sufficient risk assessment is conducted and increased risk management measures are put in place to mitigate the relevant risks.
The guidelines reference the need to account for and manage conflicts of interest that may be caused by both external and intragroup outsourcing arrangements (e.g. para 38 and recitals 26 and 37). For intragroup arrangements, we assume that this conflict could arise from service relationships between the back office (service provider) and the front office (client or outsourcing party).
In line with our response to Question 2, the guidelines’ current language can be read to imply that outsourcing between entities within the same group may create special conflicts of interest which may not already be addressed under existing regulations or processes, and may require additional measures.
It is currently unclear what specific, unique scenarios and conflicts the EBA envisions within these intragroup arrangements – which are conducted at arm’s length – which may require going beyond existing governance models and adherence to existing regulations. Additional clarity on this point would be appreciated.
We encourage the EBA to maintain a risk sensitive approach combined with the principle of proportionality throughout the final guidelines. This will be important from resourcing and burden standpoints for both supervisors and institutions, in order to ensure that outsourcing arrangements of highest importance are provided the necessary, enhanced attention and oversight.
We therefore agree with the need for enhanced documentation for outsourcing of critical or important functions, but believe the scope of covered outsourcing arrangements should rely on the full assessment criteria set out in section 9.1. To align the final guidelines with this objective, we recommend paragraph 47(c) be amended as follows in order to only apply to the outsourcing of critical or important functions:
‘in addition, the register should include at least the following information with regard to the outsourcing of critical or important functions’
Outsourcing to cloud service providers that are deemed critical or important would still be captured, based on the assessment criteria set out in section 9.1. This would however avoid an unnecessary increase in compliance burden for cloud services which do not require the application of enhanced requirements or scrutiny of both supervisor and institution.
This approach would also align with the EBA’s intention to adopt a technological neutral and proportionate approach to regulation, as outlined in its FinTech Roadmap from 15 March 2018 . Of the three approaches outlined in the Roadmap, the first included reviewing existing EU measures and the ongoing monitoring of supervisory guidance.
We support the EBA’s aim of promoting technological-neutrality in regulatory and supervisory practices and believe it should be applied wherever possible in the final outsourcing guidelines.
It is unclear what is expected from an outsourcing institution when considering whether a service provider is part of the institution’s accounting consolidated group, as highlighted in paragraph 48(f), within the context of the pre-outsourcing analysis. It would be helpful to better understand what is required in practice if a service provider belongs to an institution’s accounting consolidation group and how this may relate or impact a different type of risk appetite.
The current wording in paragraph 50 would benefit from refinement to better align the determination of a critical or important outsourcing with the full criteria set out in section 9.1, and reflect the relevant materiality of a specific arrangement.
The current approach unfortunately applies an overly broad lens where any potential activity, process or service (e.g. operational or administrative) that is potentially linked to a core business line or critical function – regardless of its relevance or materiality to the supported function – would be considered critical or important and be subject to the enhanced requirements of the guidelines. This approach ignores any sort of assessment of actual risk / importance and would result in the diversion of internal risk management and oversight resources to activities, processes or services with relatively low risks, as well as introduce unnecessary compliance burdens (e.g. enhanced documentation requirements) and costs to the outsourcing institution.
The EBA should instead focus on the full criteria that should be considered in determining if an outsourcing arrangement is critical or important and better reflect the relevance of the outsourcing. We therefore recommend paragraph 50 be amended as follows:
‘In the case of institutions, particular attention should be given to the assessment of the criticality or importance, when outsourcing activities, processes or services related to essential for the outsourcing institution’s core business lines and critical functions as defined in Article 2(1)(35) and 2(1)(36) of Directive 2014/59/EU and identified by institutions using the criteria in Articles 7 and 8 of Commission Delegated Regulation (EU) 2016/778.’
Regarding the specific assessment criteria included in paragraph 51(g) regarding ‘scaling service consumption’, we recommend that the final guidelines clarify that the assessment of scaling should specifically focus on unexpected changes to the contract which may result in a significant risk increase in connection with the outsourcing arrangement.
For example, scaling would not be a concern for outsourced services which operate on volume-based contracts and can in theory increase infinitum. Typically, the variation in these services cannot be predicted up front but are expected during the lifetime of the arrangement to occur. This is also true for outsourcing arrangements which are intended to be elastic. Service consumption flexibility as such does not in practice constitute a risk factor, as both parties usually work with volume models to cover expected costs as well as delivery capability. This actually mitigates risks for the institution as it allows more cost controls and service flexibility to better react to changes in demand and markets.
A more suitable approach would be to focus on the potential risk increase from significant new volume or importance of the service without another contract review. We therefore recommend the EBA amend the language in paragraph 51(g) as follows:
‘the possibility of the proposed outsourcing arrangement to be scaled up at the discretion of either party in an unintended manner which would result in a significant increase in risk concentration to the outsourcing institution without replacing or revising the underlying agreement;’
No comments on this section.
The expectations regarding an institution’s own risk assessment are sufficiently clear. However, certain considerations in the guidelines related to the political stability and security situation of the service provider’s jurisdiction are overly prescriptive.
Specifically, the detailed requirements set out in paragraph 61(d) ii and iii would significantly increase legal costs and the time to market for outsourcing arrangements, without providing a material benefit to the overall risk assessment. This would require for example the provision of legal memorandums for every jurisdiction where a service provider is under consideration. While political stability and security is typical part of the vendor location risk, these requirements go beyond that risk focus.
In particular, the requirements in 61(d) iii, as well as paragraph 64(h) and (i), to consider ‘any constraints’ provides limited value even within Europe, where these rights can already be challenged under European insolvency laws by the insolvency receiver. Even EU insolvency laws will prevail over any bank regulation and therefore no vendor due diligence or any clause in an outsourcing service contract is able to change mandatory rights of an insolvency receiver, e.g. challenging existing contracts and payments in accordance with available rights.
Therefore, in order to more accurately reflect these insolvency scenarios, we recommend the EBA avoid prescriptive requirements in this area by removing items i-iii under paragraph 61(d) and paragraphs 64(h) and (i).
In order to avoid restricting the further development and use of electronic signatures, greater clarity on what can be considered a ‘written agreement’ in paragraph 62 would be beneficial. In particular, alignment with the Regulation on electronic identification and trust services for electronic transactions in the internal market (eIDAS Regulation) should be secured to confirm the acceptance of advanced electronic signatures and further drive electronic execution of contracts in line with the technological progress in the area of digital contracting. This is increasingly important as the use of advanced electronic signatures will be a necessary element to satisfy secure electronic contracting processes going forward with a reasonable amount of implementation effort and sufficient technical security.
Conversely, any requirements for qualified electronic signatures for all outsourcing contract types would create significant burdens on the implementation of digital processes and should not be considered as proportional, in consideration of the intention of the eIDAS Regulation.
We therefore recommend the addition of a new definition for ‘written agreement’ under paragraph 11, as follows:
‘means any contract document signed by authorised representatives of the involved parties either physically in writing or electronically, using advanced electronic signatures in accordance with the EU Regulation No 910/2014 on electronic identification and trust services for electronic transactions in the internal market (eIDAS Regulation)’
The requirements set out in paragraphs 63(h), 66(b) and 72(b) for the outsourcing institution to secure ‘unrestricted’ rights to access and audit the service provider and subcontractors for any outsourcing relationship will pose significant legal and practical challenges, as well as security concerns for both the outsourcing institution and service provider.
This will be particularly the case for providers of standardised services (e.g. cloud service providers) which also have a significant number of customers. While we fully agree that access and audit rights are essential for the monitoring and supervision of critical outsourcing arrangements, requiring unrestricted access to any outsourcing arrangement would unnecessarily overreach, introduce new risks to both sides of the outsourcing arrangement and potentially prevent the adoption of existing and potentially new outsourcing arrangements in the developing financial services ecosystem.
This requirement also exceeds approaches employed by regulators in other jurisdictions, which require securing a necessary or suitable right to access, to allow for appropriate supervision of the outsourced activity. Policymakers and supervisors are increasingly adapting their frameworks to reflect these challenges in the face of technological and business model changes.
For example, the UK Financial Conduct Authority (FCA) requires that ‘the firm, its auditors, the FCA and any other relevant competent authority must have effective access to data related to the outsourced activities, as well as to the business premises of the service provider; and the FCA and any other relevant competent authority must be able to exercise those rights of access’ . This same ‘effective access’ approach is included in Article 31(2)(i) of the MiFID Org Regulation.
Similarly the U.S. Office of the Comptroller of the Currency’s (OCC) Risk Management Guidance requires that the bank ‘ensure that the contract establishes the bank’s right to audit, monitor performance, and require remediation when issues are identified’ .
Similar, more flexible approaches are employed in both the U.S. and APAC, which do not impose such unconditional and potentially inoperable requirements for outsourcing arrangements (see Annex 2 for examples of specific regulatory language).
Amending the language in the EBA guidelines to achieve a more feasible and proportionate standard is especially important, as this will apply to not only all outsourcing arrangements, but also any sub-outsourcing. A consistent approach across jurisdictions will also better support the smooth functioning and central risk management of outsourced arrangements for large, globally active financial institutions and the scaling of innovative solutions in a responsible and controlled manner.
We therefore recommended that paragraph 63(h) be amended as follows:
‘the necessary and effective right of institutions, payment institutions and competent authorities to get any information needed with regard to the outsourcing and to access and audit the service provider as further specified in Section 10.3;’
Paragraph 72(b) would similarly be amended as follows:
‘necessary rights of inspection and auditing related to the outsourcing arrangement (‘audit rights’), to enable them to effectively monitor the outsourcing arrangement and to comply with all applicable regulatory requirements.’
The guidelines also introduce a significant, disproportionate change to the current approach on non-critical outsourcing requirements by requiring specific contract content for these non-critical services without respecting an informed, risk-based view. It is especially challenging to expect unlimited or unrestricted audit and inspection rights in a contract with a non-critical service provider. This issue is further complicated by the broad outsourcing definition applied in the guidelines – which may now include all kinds of activities to be covered by an extensive outsourcing arrangement – instead of a risk-based content.
Therefore, we recommend to at least limit the audit clause requirements of paragraph 63 (g) and (h) as well as Section 10.3 to critical and important functions, and otherwise refer the applicability for non-critical services to the extent required by the institution for its effective risk management.
We also encourage the EBA to consider a more proportionate and forward-looking approach on sub-outsourcing / chain outsourcing. As highlighted in our response to the EBA’s recommendations on cloud outsourcing in 2017, it is extremely difficult for a financial institution to have control of a cloud service provider’s whole outsourcing chain, due to the more dynamic nature of the cloud environment and larger volume of customers than what is found in traditional outsourcing environments.
We therefore recommend the EBA to accept alternatives to direct firm oversight which would more effectively address the identified risks and nature of dynamic sub-outsourcing. For example, in a recent report released by U.S. Treasury on regulatory reforms to promote innovation , the Treasury recommended that supervisors formally recognise independent audit and security standards that sufficiently meet regulatory expectations and set clear and appropriately tailored expectations for chain outsourcing. This was done in order to provide for a ‘prudent and informed migration of activities to the cloud’. The adoption of such alternative solutions and oversight mechanisms for compliance of sub-outsourcing, such as the requirements in paragraph 66 and the due diligence standards set out in section 9.2, would better enable the adoption of cloud services for financial institutions and better future-proof the guidelines to changes in technology and the financial services landscape.
The requirements in paragraph 76 set out the need for the outsourcing institution to ensure its ability to carry our security penetration testing to assess the effectives of implemented cyber and internal ICT security control and mitigation measures. This is an important area of risk management and will be increasingly so as the financial services sector continues its current wave of digital transformation.
However, this requirement can be read to require that each outsourcing institution secure the right to conduct their own individual penetration testing (whether by internal staff or external experts) on the service provider’s systems.
While the requirement to conduct penetration testing is not a new concept or standard, the potential application of such a narrow approach where only the outsourcing institution can conduct the testing may be inoperable in practice and create extensive security issues for both the outsourcing institution and service provider. This again is especially true for standardised services such as cloud services. For example, a scenario where every customer of a large cloud service provider – whose customers can number in the hundreds of thousands – is entitled to conduct penetration testing would create significant operational complexity and increase risks of data exposure, corruption, theft and leakage, as well as potential denial or disruption of service and performance issues for other customers.
A more secure and workable approach would be to clarify that the outsourcing institution can meet the requirement for security penetration testing through a number of methods. Current best practice includes offering alternatives to direct client testing, such as acceptance of reports from reputable third parties or to offer assistance to jointly conduct the testing alongside the service provider. This would be in the same spirit as the flexibility provided in paragraph 75 for pooled audits, and similarly decrease the operational burden on, and risks for, both the client and the service provider. The language in paragraph 76 should be amended to reflect such an approach as follows (edits in bold):
‘In line with the EBA Guidelines on ICT Risk Assessment under the Supervisory Review and Evaluation process institutions should, where relevant, ensure the ability to carry out security penetration testing to assess the effectiveness of implemented cyber and internal ICT security measures and processes. Taking into account Title I, payment institutions should also have internal ICT control mechanisms, including ICT security control and mitigation measures. The outsourcing institution may meet this requirement by acquiring reasonable evidence on the security, either directly or by the vendor, through an external tester, and perform the test under their control.’
No comments on this section.
No comments on this section.
We agree with the need to harmonise requirements across markets participants but recommend the EBA consider industry best practices for the process and manner of reporting material outsourced activities to competent authorities.
For example, under the German Banking Act, the outsourcing institution includes information on outsourcing activities in its audit reports. This arrangement has worked well to date in Germany for both the financial institution and regulator in carrying out appropriate risk management and supervisory tasks; a separate or standalone notification, which was previously utilised, has not been deemed necessary or efficient for the accomplishment of these tasks.
This ex-post approach still provides the right of the supervisor to request unwinding (or amendments of) any outsourcing arrangement that would not comply with the national rules set up under the guidelines. Furthermore, certain information required for notification is usually not available until after negotiations have been finalised, e.g. outsourcing agreement reference number. In other situations, information might only be available if negotiations are in a very advanced stage.
The need for ex-ante notification would not only increase the time to market of outsourcing arrangements and add additional operational burdens and costs, but would do so without providing additional value to the supervisor looking at the overall strategy of an institution, including generally implemented risk control functions.
A more efficient approach to ensuring appropriate due diligence and assessments are completed prior to proceeding with an outsourcing arrangement, would be for regulators to focus on the robustness of an institution’s internal governance and control frameworks. Assessing an institution’s processes would address the execution of appropriate risk management practices for outsourced activities, which are necessary for adapting to the development of new and more complex levels of services and technologies, e.g. in the cloud environment.
In short, focusing on process and governance would be more effective as process equals outcome. This would better position the EBA and local supervisors to account for the rapidly evolving technological landscape, subsequent changes in outsourcing arrangements and potential new providers.
However, if the EBA is intent on mandating ex-ante notification, this should only require a brief summary of the outsourcing arrangement and should not be subject to pre-authorisation from the competent authority.