# SMSB Capital and Liquidity Requirements - Consultative Document

## Document Properties

• Type of Publication: Consultative Document
• Date: January 2020

## 1.0 Introduction

OSFI's mandate is to protect depositors, policyholders and creditors by developing a regulatory framework to manage and mitigate risk, assessing the safety and soundness of financial institutions and intervening promptly when corrective actions are needed. OSFI recognizes the importance of allowing financial institutions to compete effectively and take reasonable risks and holds boards and senior management ultimately responsible for the viability of their institutions.

In July 2019, OSFI released a discussion paper: Advancing Proportionality: Tailoring Capital and Liquidity Requirements for Small and Medium-Sized Deposit-Taking InstitutionsFootnote 1 (July 2019 Discussion Paper) which described its initiative to develop more tailored requirements that take into account the unique nature of these institutions. As a discussion paper, the document aimed to solicit input from stakeholders at an early stage of the policy development process, enhance OSFI's transparency to external stakeholders, and improve Canadians' understanding of the objectives guiding our work. It outlined specific proposals for how OSFI could achieve greater proportionality through a combination of small and medium-sized deposit-taking institutions (SMSB) segmentation and more selective use of capital and liquidity measures, and sought feedback in response to specific proposals.

The purpose of this consultative document is to provide stakeholders with an overview of feedback that was received in response to the July 2019 Discussion Paper and an update regarding the development of the Pillar 1Footnote 2 SMSB capital and liquidity frameworks. The document outlines changes which are proposed to the framework, and seeks further feedback from stakeholders in advance of the publication of draft SMSB capital and liquidity requirements in late spring 2020. Future phases of this initiative which relate to Pillar 2 and Pillar 3 capital and liquidity requirements will be the subject of consultations later this year.Footnote 3

The proposals in this consultative document aim to strike an appropriate balance in relation to the following principles which are guiding OSFI's review of the capital and liquidity frameworks for SMSBs.

1. The capital and liquidity frameworks applied to SMSBs should reflect the nature, size, complexity and business activities of these institutions.
2. Capital and liquidity requirements should contribute to the protection of depositors and creditors and allow institutions to compete effectively and take reasonable risks.
3. The revisions to the frameworks should strike the right balance between improving the risk sensitivity of the requirements for SMSBs and reducing the complexity of the frameworks to make them more fit for purpose.

### 1.1 General feedback received in response to the July 2019 Discussion Paper

In September 2019, we received comments from stakeholders on the proposals outlined in the July 2019 Discussion Paper. The main overarching themes that were raised by stakeholders are summarized below, while specific themes that focused on a particular topic (e.g., credit or operational risk capital, and liquidity requirements), are addressed within subsequent sections of this document.

#### Overarching themes noted in consultation responses to the July 2019 Discussion Paper

• Respondents expressed broad support for the initiative to advance proportionality, specifically with respect to the introduction of a segmentation approach and tailoring Pillar 1 requirements for each category of institutions.
• Many respondents expressed support for additional risk sensitivity within the capital and liquidity frameworks, even if this were to contribute to some increased administrative complexity. There was, however, also considerable interest among smaller SMSBs regarding the potential simplification of certain requirements outlined in the proposals (e.g., a simplified standardized approach for credit risk, a flat rate add-on for operational risk, and simplified liquidity metrics).
• Some respondents expressed concerns that higher capital and liquidity requirements could be required as a trade off for a simplified framework. Most of these respondents expressed a preference for maintaining the current set of requirements if this were to be the case.
• SMSBs emphasized the importance that OSFI's proportionality work also include Pillar 2 capital and liquidity guidelines, supervisory practices (e.g., ICAAP), as well as Pillar 3 disclosure requirements.
• Several respondents, including many smaller institutions, stressed the importance of having a compliance and reporting regime that is as simple and cost effective as possible. Respondents also noted that changes to the capital and liquidity regime should take into consideration the resources that will be required to implement them as well as the ongoing burden they could place on institutions.
• Respondents requested that OSFI undertake additional consultation when the SMSB capital and liquidity framework proposals are more fully developed in order to be able to evaluate potential implications arising from prospective changes.
• Some institutions suggested that encouraging competition and innovation in the Canadian banking environment should be the priority when considering changes to the capital and liquidity regime for SMSBs.

In response to the overarching themes, OSFI has the following comments:

• As noted above, one of the guiding principles for this initiative is to strike the right balance between improving the risk sensitivity of the requirements for SMSBs and reducing their complexity. OSFI's initiative to develop more tailored capital and liquidity requirements does not have an objective of increasing the level of capital and liquidity held by institutions across the industry. However, in seeking to improve risk sensitivity or reduce complexity, the implication may be that certain institutions are more affected by the proposed changes which may lead in some cases to higher Pillar 1 capital or liquidity requirements.
• OSFI acknowledges the importance of reviewing Pillar 2 and 3 capital and liquidity requirements as part of this initiative. However, given the foundational nature of the Pillar 1 requirements, OSFI felt that this was an appropriate starting point. Consultation with respect to Pillar 2 and 3 requirements is expected to begin later this year.
• OSFI understands that institutions require sufficient time to implement changes to the capital and liquidity frameworks for SMSBs. The consultation strategy related to the SMSB capital and liquidity frameworks includes several milestones (i.e., discussion paper, consultative document and draft guidelines) to allow for feedback throughout the policy development process and provide an early indication of potential new requirements which should assist institutions to consider and plan for the changes. OSFI is targeting publication of a final set of rules by December 2020 in order to provide institutions approximately one year of final preparation for the changes.
• This consultative document is part of OSFI's consultation strategy and provides stakeholders with more details regarding the proposed capital and liquidity requirements as well as tools to quantify the potential impact for certain aspects of the revised SMSB capital and liquidity frameworks. In addition to the consultative document, OSFI plans to issue draft capital and liquidity guidelines in late spring 2020 which will provide stakeholders with the proposed detailed requirements related to the revised frameworks.
• With respect to the suggestion that OSFI should prioritize the encouragement of competition and innovation when considering changes to the capital and liquidity frameworks, OSFI notes that there are three principles guiding the review of the capital and liquidity frameworks for SMSBs. It would not be appropriate to prioritize certain aspects of the principles to guide our work on this initiative. Doing so could also be inconsistent with our mandate. Instead, OSFI is seeking to strike an appropriate balance between these principles in considering changes to SMSBs capital and liquidity requirements.

The remainder of this document outlines more specific feedback that was received and outlines updated proposals in relation to five key areas of the framework. Details related to how the frameworks could be operationalized are also provided. Each section concludes with a series of questions to interested stakeholders.

## 2.0 Segmentation

In the July 2019 Discussion Paper, OSFI proposed segmenting SMSBs into distinct categories which would allow for more tailoring of the Pillar 1 capital and liquidity requirements in line with the guiding principles outlined in the same document. The proposed segmentation was based on a combination of four qualitative and quantitative criteria as follows:

• The approach used to calculate Pillar 1 capital requirements for credit risk;
• The size of an institution's on-balance sheet assets;
• The amount of assets under administration (AUA) / assets under management (AUM); and;
• OSFI judgement.

### 2.1 Specific feedback received on the July 2019 Discussion Paper (Segmentation)

Several stakeholders provided feedback regarding the segmentation proposal. Overall, respondents were supportive of the initiative to segment SMSBs. However, certain respondents felt that the categorization outlined in the July 2019 Discussion Paper could be further refined.

First, it was noted that the main differentiator between categories was based on size and the criteria did not sufficiently consider the nature of SMSBs' activities. Second, given the current segmentation criteria, the total asset threshold of \$500 million between Category 3 and 4 was considered too low. Third, several respondents felt that having a separate Category 1 based on whether an SMSB had been approved to use the internal-ratings based (IRB) approach for credit risk provided little incremental value as, apart from the Net Stable Funding Ratio (NSFR), the other requirements were the same as for Category 2 institutions.

Stakeholders also requested additional details related to the operationalization of category migration. In particular, more details were requested regarding OSFI's criteria for the use of its own judgement as part of the segmentation and category migration. Finally, certain respondents suggested that SMSBs should have the option to move between categories with OSFI approval only being required when an institution is seeking to move to a lower category (e.g., from Category 3 to Category 4).

### 2.2 Proposed changes to segmentation criteria

OSFI has considered the feedback received during the consultation period and is proposing the following changes to the segmentation criteria.

#### Category 1 and 2 Segmentation

As discussed in the Liquidity Risk section of this consultative document, OSFI is proposing that the applicability of the NSFR be linked to the level of wholesale funding reliance for SMSBs with total assets greater than or equal to \$10 billion. Given this change, the remaining capital and liquidity requirements for Category 1 and 2 institutions are the same except that Category 1 institutions have been approved to use the IRB approach for credit risk. As such, we believe it would be beneficial to merge Category 1 and 2 to reduce duplication and to reflect the similarities among these medium-size institutions. Going forward, we will refer to this new merged category as Category I or "Medium-sized Institutions".

#### Category 3 and 4 Segmentation

As noted above, one of the main feedback items received through the consultation was that the segmentation criteria did not sufficiently consider the nature of SMSBs' activities. This is particularly relevant to Category 3 and 4 institutions that have total assets of less than \$10 billion and represent 56 of the 63 SMSBs. When considering the nature of the institutions within these categories, a primary distinction relates to whether or not their main business activities involve financial intermediation. SMSBs involved in financial intermediation are primarily conducting lending activities.

SMSBs that conduct lending activities generally require more leverage and funding sources in order to conduct their business than institutions focused on non-lending activities. For these SMSBs, it will be important to have a capital measure for leverage (i.e., the leverage ratio) and a risk sensitive measure for credit risk and operational risk (i.e., the risk-based capital ratio). Given the liquidity risk associated with their financial intermediation activities, it will also be important to have both a short term measure (i.e., the Liquidity Coverage Ratio or LCR) and a longer term measure (i.e., a cash flow metric).

SMSBs that are not involved in financial intermediation generally have lower leverage, fewer funding / liquidity requirements and conduct activities for which operational risk is the more prominent risk (e.g., fee based trust and custody businesses). For these SMSBs, it will be important to have a capital measure that captures operational risk, while the liquidity measure could be simpler given their cash flow requirements.

In order to segment Category 3 and 4 institutions based on business activity, OSFI considered the following measures to assess whether an institution is involved in financial intermediation: (i) total loans on balance sheet, and (ii) total deposits. Total loans is considered a more comprehensive measure as it provides an indication of the level of potential credit risk and leverage an SMSB may have. As well, SMSBs may use different funding sources for their financial intermediation activities and total deposits would not capture all of these.

As such, OSFI is proposing to change the segmentation criteria for Category 3 and 4 institutions from total assets and AUA/AUM to total loans. The minimum threshold would be set at \$100 million in total loans and is based on an analysis of the dollar amount of loans held by SMSBs where lending is a significant activity.

Under this proposal, Category 3 would include small lending institutions where total loans are greater than \$100 million and total assets are less than \$10 billion. Going forward, we will refer to Category 3 as Category II or "Small Lenders". Category 4 institutions would consist of non-lending institutions that are primarily involved in fee based activities such as trust and custody businesses. Going forward, we will refer to Category 4 as Category III or "Non-Lenders".Footnote 4

Table 1 provides a comparison of the segmentation criteria in the July 2019 Discussion Paper and the new proposed segmentation criteria.

Table 1
July 2019 Discussion Paper proposal Jan 2020 Consultative Document proposal
Category 1 Approved to use IRB approach for credit risk Medium-sized Institutions (Category I) Assets > \$10B
Category 2 Assets > \$10B
Category 3 Assets of \$0.5B to \$10B or AUA/AUM > \$20B Small Lenders (Category II)

Assets <\$10B and
Total Loans > \$100M

Category 4 Assets < \$0.5B and AUA/AUM < \$20B Non-Lenders (Category III)

Assets <\$10B and
Total Loans < \$100M

Table 2 provides a comparison of the number of institutions by category as well as the characteristics of institutions by category based on the new proposed segmentation criteria.

Table 2
Categories based on July 2019 Discussion Paper Composition based on July 2019 Discussion Paper Segmentation Categories based on Jan 2020 Consultative Document Composition based on Jan 2020 Consultative Document Segmentation Characteristics based on Jan 2020 Consultative Document Segmentation
Category 1 1 institution representing 30% of total SMSB assets Medium-sized Institutions (Category I) 7 institutions representing 76% of total SMSB assets Medium-size institutions with operations ranging from diverse (e.g., multiple business lines, multiple distribution channels) to relatively less complex and less diversified.
Category 2 6 institutions representing 46% of total SMSB assets
Category 3 37 institutions representing 23% of total SMSB assets Small Lenders (Category II) 36 institutions representing 23 % of total SMSB assets Smaller (based on asset size) and generally less complex institutions primarily focused on lending activities.
Category 4 22 institutions representing 1% of total SMSB assets Non-Lenders (Category III) 20 institutions representing 1 % of total SMSB assets Smaller (based on asset size) and generally less complex institutions primarily focused on non-lending activities.

OSFI believes that the proposed changes to the segmentation criteria provide the following benefits:

• Ensures the segmentation criteria is more activity / risk based.
• Greater consistency in requirements for institutions with similar business activities and risks, which also facilitates comparisons amongst institutions.
• Less expected migration between categories compared to the segmentation criteria outlined in the July 2019 Discussion Paper.

### 2.3 Migration between categories and use of OSFI judgement

As part of the July 2019 Discussion Paper consultation process, certain stakeholders suggested that SMSBs should have the ability to choose their categorization, but that this be subject to OSFI approval if an SMSB was seeking to move to a lower category (e.g., moving from Category 3 to Category 4).

OSFI has considered this feedback; however, we do not believe this type of optionality is beneficial for the SMSB capital and liquidity framework. Indeed, the proposed changes to the segmentation criteria noted above ensure institutions with similar business activities have comparable Pillar 1 capital and liquidity requirements. Moreover, limiting the amount of optionality in the framework makes it simpler to understand and easier to administer. As such, OSFI is proposing that SMSBs not have the option to migrate to another category if they do not meet the segmentation criteria.

In limited circumstances, OSFI will retain the discretion to move an institution into a different category based on OSFI's own judgement. Following are some of the factors that OSFI would consider in re-categorizing an institution:

• Forward looking considerations that would materially shift an institution's business activities. For example, a Non-Lender may decide to change strategic direction and focus on financial intermediation activities through lending. Under these circumstances, OSFI may decide to move the SMSB to the Small Lender category before it reaches the \$100 million total loan threshold. Also an SMSB that is in the process of winding down its business could warrant a change in category depending on the risk remaining in its operations.
• The nature of an SMSB's business activities / model warrants a change in category. For example, an institution that requires leverage as part of its business model but does not do lending. Under these circumstances, OSFI may decide to move the SMSB from the Non-Lender category to the Small Lender category.

### 2.4 Operationalization of the segmentation

In order to operationalize the segmentation process, an institution's total assets and total loans would be calculated based on the average of the amount reported in the M4 monthly Balance Sheet return for the 12 months in the institution's previous fiscal yearFootnote 5. If an institution crosses a threshold, they would be given a year to implement the requirements of their new category.

For the initial implementation in Q1 2022, the threshold calculation would be performed based on total assets and total loans from fiscal 2020 (using month-end data for all of 2020).

After implementation, a comparison of an institution's total assets and/or total loans against the threshold criteria would be required on an annual basis. The threshold calculation would be based on an average of monthly data for the previous fiscal year. Institutions that have crossed a threshold would be given a year to implement the requirements of their new category. For example, if a Non-Lender's average total loans crosses the threshold (by going above \$100 million), the institution would be expected to meet the requirements for the Small Lender category effective Q1 of the following year. Once an institution migrates to a new category, it would be expected to remain in the category for a minimum of two fiscal years. This will provide institutions with more certainty regarding Pillar 1 capital and liquidity requirements over a set time period rather than the potential for changes on an annual basis if an institution is operating close to a threshold. If after two fiscal years an institution crosses one of the segmentation criteria thresholds, they would be given a year to implement the requirements of their new category.

The following illustrates how the threshold for segmentation would operate. The example focuses on the migration between the Small Lender and Non-Lender categories (however the process would be the same with the other category as well).

• For Q1 2022, the total loans threshold would be assessed using fiscal 2020 data. If the average total loans using fiscal 2020 data is above \$100 million, the institution has crossed the Small Lender Category criteria threshold and would need to meet the capital and liquidity requirements for the Small Lender category for fiscal years 2022 and 2023.
• In Q1 2023, the calculation would be performed again using fiscal 2022 data. If the average total loans using fiscal 2022 data is below the \$100 million threshold, the institution has crossed the Non-Lender Category criteria threshold and would need to meet the capital and liquidity requirements for the Non-Lender Category for fiscal year 2024 and 2025.

In the case of OSFI judgement, institutions would typically be required to meet the new requirements in the quarter ending one year after notification. For example, if an institution in the Non-Lender Category is notified in Q2 2023 that OSFI will be moving it to the Small Lender Category based on its judgement, the institution would need to implement the Small Lender Category capital and liquidity requirements starting in Q2 2024.

### 2.5 Questions on the segmentation proposals

1. Do you have any comments regarding the new proposed segmentation criteria including the criteria thresholds?
2. Do you have any comments regarding the operationalization of the segmentation process?

## 3.0 Credit Risk

The July 2019 Discussion Paper proposed a credit risk framework that aimed to better capture the risks faced by SMSBs while reducing complexity. The paper suggested that this framework could include a more risk sensitive standardized approach (currently being developed pursuant to the Basel III reforms), while maintaining a simpler standardized approach (based on the 2019 CAR Guideline). Institutions with more than \$10B in assets (Category 2) would be subject to the Basel III standardized approach with the possible option of using a simpler approach. Category 3 institutions would be subject to a simplified standardized approach but may be given the option to apply the more risk sensitive standardized approach.

### 3.1 Specific feedback received on the July 2019 Discussion Paper (Credit Risk)

In the comments received on the July 2019 Discussion Paper, many respondents viewed the greater risk sensitivity of the revised standardized approach for credit risk (based on Basel III) very favourably despite the additional complexity that will be associated with this approach. Most of these respondents attributed these views to an expectation that additional risk sensitivity of credit risk requirements would allow them to compete more effectively with other institutions. However, some respondents expressed concerns regarding operational challenges that could be faced by institutions with respect to new data requirements and reporting capabilities.

Respondents noted that retaining a simpler standardized approach based on the 2019 CAR Guideline would not enable Category 3 institutions to benefit from the improved risk sensitivity of the standardized approach. It could also create a gap in capital requirements between institutions in Categories 2 and 3 which could have competitive implications. Some respondents also recommended that OSFI examine the treatment of asset classes where there is a noteworthy differential between the risk weights applied under the standardized approach and those resulting from the IRB approach.

### 3.2 Proposed changes to the credit risk proposals

In response to industry comments and as a result of further internal analysis, rather than having two separate standardized approaches to credit risk, OSFI is proposing to have a single standardized approach. This approach will include the more risk sensitive treatment as well as a simplified, less granular, treatment available for immaterial asset classes (i.e., those with less than \$200 million in total exposure).

In contrast to what was proposed in the July 2019 Discussion Paper, Medium-sized Institutions will be required to apply the more risk sensitive treatment to all of their assets; the simplified treatment will not be available.

For Small Lenders (Category II), the revised standardized approach will operate as follows:

• For some asset classes (see below), there will only be one treatment available (i.e., there will be no distinction between the more risk sensitive treatment and the simplified treatment).
• For other asset classes there will be two treatments available. Institutions will be required to apply the more risk sensitive treatment to their material asset classes and will have the option to apply the simplified treatment to their immaterial asset classes.
• For activities such as derivatives and securitization, there will be no simplified treatment available. Institutions will be required to use the more risk sensitive treatment if they are engaged in these activities.

For the following asset classes there will only be one treatment available under the revised standardized approach. OSFI regards the standardized treatment as sufficiently simple that the development of a more simplified treatment for these asset classes is not warranted.Footnote 6

• Sovereigns and central banks – exposures to national governments and the independent national authority that acts as the financial and banking agent for the national government.
• Public sector entitiesexposures to entities directly and wholly-owned by a government (federal, provincial, or municipal).
• Multilateral development banks – exposures to institutions created by a group of countries, who act as joint owners, but with independent legal and operational status to provide financing and professional advice for economic and social development projects.
• Land acquisition, development and construction loans to companies or Special Purpose Vehicles financing the acquisition of land for development and construction, for which repayment depends on the future sale or lease of the property.
• Reverse mortgages – non-recourse loans secured by property, that have no defined term and no monthly repayment of principal and interest, for which the amount owing grows with time (as interest is accrued and deferred) and is generally repaid from the net proceeds of the sale of the property (i.e., net of disposition costs) after the borrower has vacated the property.
• Non-qualifying retail exposures to individuals or persons that do not meet the specified criteria for retail exposures.
• Subordinated debt, equity and other capital instrumentsinvestments in instruments issued by either corporates or other institutions and that are not deducted from regulatory capital.
• Equity investments in fundsequity investments in all types of funds held in the banking book, including off-balance sheet exposures.
• Other assets – all assets not covered under a specific asset class in the standardized approach.

For the following asset classes, a simplified treatment will be available. The following groupings of asset classes should be used to determine whether an asset class is material (where possible, the associated data points from the 2019 BCAR regulatory return have been identified for reference):

1. Banks, securities firms and other financials treated as banks (including covered bonds issued by these institutions) (BCAR Schedule 8, measure M11, Total (500))
2. CorporatesFootnote 7, SMEs treated as Corporates, securities firms and other financials treated as Corporates, and specialised lending (Project Finance, Object Finance, Commodity Finance) (BCAR Schedule 5, measure M11, Total (500))
3. Commercial real estateFootnote 8
4. Residential real estate (including Home Equity Lines of Credit) (BCAR Schedule 9, measure M11, Total (500))
5. Qualifying retail (including credit cards and auto loans) and Small Business Entity retail (BCAR Schedules 10 and 11, measure M11, Total (500))

Under the simplified treatment, institutions would not be required to split these groupings into sub-asset classes. For example, all Corporate exposures including specialised lending would be grouped together and the same treatment applied. Institutions would not be required to assess the exposure against the specialised lending criteria to determine if a separate treatment should apply.

OSFI believes the proposed changes to the standardized approach will help to improve the risk sensitivity of this approach. This includes the revisions reflected in the Basel III reformsFootnote 9 as well as domestic modifications made by OSFI to reflect the Canadian market and in response to feedback from stakeholders. Subsequent to the implementation of these changes, OSFI will consider whether additional changes to the standardized approach are needed and would consult on any proposed changes as part of our normal guidance development process.

### 3.3 Operationalization of the material asset class threshold

Material asset classes are those with greater than \$200 million in total exposuresFootnote 10. The exposure amount would be based on an average of the end-of quarter amounts calculated at fiscal year end using datapoints from the BCAR regulatory return.

For the initial implementation in Q1 2022, the threshold calculation would be performed based on BCAR data from fiscal 2020 (using quarter-end data from Q1, Q2, Q3 and Q4 of 2020).

After implementation, the threshold calculation would be performed on an annual basis. The calculation would be based on an average of fiscal quarter-end BCAR data. If a Small Lender's position relative to the thresholds has changed from the previous year, the institution would be given a year to implement the applicable treatment. For example, if an asset class that was considered material becomes immaterial (by falling below the \$200M threshold), the Small Lender has the option to use the simplified treatment for the asset class effective Q1 of the following year. Conversely, if an asset class that was previously considered immaterial becomes material (by rising above the \$200M threshold), the Small Lender would be required to use the more risk sensitive treatment for the asset class effective Q1 of the following year. In addition, to ensure some stability in the capital treatment, once a Small Lender treats a portfolio as material or immaterial, it would be required to maintain that treatment for two years.

The following example illustrates how the threshold for a material asset class would operate.

For Q1 2022, the threshold for Corporate exposures would be assessed using fiscal 2020 data:

Table 3
BCAR Schedule 5
(Measure M11, Total 500)
Q1 2020 Q2 2020 Q3 2020 Q4 2020 Average
\$210M \$205M \$207M \$215M \$209M

Since the average using fiscal 2020 data is above \$200 million, the Small Lender's Corporate exposures would be deemed material and capital requirements would need to be calculated using the more risk sensitive treatment for fiscal years 2022 and 2023.

In Q1 2023, the calculation would be performed again using fiscal 2022 data:

Table 4
BCAR Schedule 5
(Measure M11, Total 500)
Q1 2022 Q2 2022 Q3 2022 Q4 2022 Average
\$190M \$180M \$185M \$200M \$189M

Since the average exposure amount is below the \$200 million threshold, the Small Lender would have the option of using the simplified treatment effective Q1 2024. If the Small Lender switched to the simplified treatment for 2024, it would be required to use this treatment for fiscal 2025 as well.

### 3.4 Availability of the internal ratings-based (IRB) approach to credit risk

As noted in Section 2, Categories 1 and 2 as described in the July 2019 Discussion Paper will be merged into a single category based on total assets. There will no longer be a distinction between Categories 1 and 2 based on IRB approval. In addition, only institutions with at least \$10 billion in assets (i.e., Medium-sized Institutions) will be permitted to apply to use the IRB approach for purposes of determining regulatory capital requirements. Institutions using the IRB approach are expected to have in-depth rating systems that meet all of the IRB minimum requirements as described in the CAR Guideline. This requires institutions to have granular loss data over a long time horizon as well as the ability to validate and back test IRB parameters. Given that the development and maintenance of these ratings systems requires significant resources (both in terms of IT systems and from a staffing perspective), the availability of the IRB approach will be limited to institutions with at least \$10 billion in assets (i.e. Medium-sized institutions). All other institutions must use the standardized approach to determine regulatory capital requirements.

### 3.5 Summary of credit risk proposals

The following table compares the credit risk treatment proposed in the July 2019 Discussion Paper to the treatment set out in this document.

Table 5
July 2019 Discussion Paper Proposal Jan 2020 Consultative Document Proposal
Category 1 IRB Approach Medium-sized Institutions
(Category I)

Basel III Standardized Approach

Option to apply to use IRB

Simplified treatment not available

Category 2 Basel III Standardized Approach
Category 3 Simplified Standardized Approach based on Basel II Small Lenders
(Category II)

Basel III Standardized Approach for material asset classes with simplified treatment for immaterial asset classes

No option to apply to use IRB

Consistent with the stated principles for OSFI's review of the capital and liquidity frameworks for SMSBs, the proposed approach to credit risk will allow for increased risk sensitivity where it is most relevant and maintain a level of simplicity where the risks and prudential benefits do not justify additional costs.

### 3.6 Questions on credit risk proposals

1. Given the description of the credit risk standardized approach, do you think it is beneficial to have a simplified treatment available for Small Lenders to use for portfolios below the \$200 million threshold?
2. Do you have any views on the asset classes for which a simplified treatment is being proposed? Are there other asset classes for which a simplified treatment should be considered and, if so, why?
3. Do you have any comments on the level and/or operationalization of the \$200 million materiality threshold?

## 4.0 Operational Risk

The July 2019 Discussion Paper proposed that Category 1 and 2 institutions would be subject to the revised Basel III standardized approach for operational risk (Revised Standardized Approach). Category 3 SMSBs would continue to use a simple approach to calculate operational risk capital requirements, which could either take the form of the existing Basic Indicator Approach (BIA) under Basel II or could be based on another measure (e.g., assets or AUA/AUM) to determine the flat rate capital add-on for operational risk. Category 4 SMSBs would not be subject to risk-based capital requirements.

### 4.1 Specific feedback received on the July 2019 Discussion Paper (Operational Risk)

In the comments received on the July 2019 Discussion Paper, some respondents indicated that having materially different operational risk capital requirements between Category 1 and 2 versus Category 3 institutions would reduce comparability between institutions, potentially disadvantaging one category over the other, depending on the final form of the flat rate add-on. It was also noted that, depending on how it was calibrated, this could lead to material changes in capital requirements due to the migration of an institution between categories. Some respondents advocated for larger SMSBs having the ability to use the more risk-sensitive Revised Standardized Approach using loss data. While one respondent supported the idea of applying the Revised Standardized Approach to all SMSBs without the use of loss data, several others supported keeping the existing BIA due to its simplicity. One respondent noted that by using gross income as an indicator, the BIA is punitive to institutions with high net interest margins, since it resulted in higher minimum operational risk capital requirements relative to similar sized institutions with lower interest margins. Another respondent noted that the BIA, counterintuitively, reduces operational risk capital requirements for an institution that realizes certain losses (e.g., trading losses).

### 4.2 Options considered for operational risk capital

For Small Lenders (Category II), OSFI considered the development of a flat-rate add-on for operational risk, such as a percentage of total assets or credit risk-weighted assets. After further analysis, this was not deemed appropriate because many SMSBs have fee-based activities that are not reflected on their balance sheets. This is particularly relevant since Non-Lenders (Category III) will also continue to have an operational risk capital requirement (see Section 5), and it would be simpler and more comparable for them to use the same approach. An add-on based on off-balance sheet activities (e.g., AUA/AUM) was also considered; however, given the heterogeneity of SMSB fee-based activities, we believe it would be difficult to appropriately calibrate a suitable add-on that provides a meaningful improvement over revenue-based indicators, especially given the added complexity this would entail. OSFI is therefore of the view that a revenue-based indicator for operational risk capital requirements remains most suitable and comparable for SMSBs across different categories with different business models and activities.

While the BIA is a simple and intuitive revenue-based measure, it has some deficiencies. As noted in the feedback received, using net interest income as a proxy for size and operational risk may be overly punitive for high-margin lending institutions. In addition, trading losses reduce the capital requirement under the BIA. The BIA also does not include realized gains or losses from the sale of assets in the banking book, and income from joint ventures in gross income, which are also related to operational risk.

The Revised Standardized Approach was designed in part to address these deficiencies in the BIAFootnote 11. It also provided greater risk-sensitivity by isolating fee and commission revenues and expenses and including internal operational risk losses in a revised definition of gross income (Business Indicator Component). To achieve this additional risk-sensitivity, however, the Revised Standardized Approach requires institutions to calculate and report line items that are not part of their financial statements or existing OSFI returnsFootnote 12. The additional prudential and risk management benefits that would result from SMSBs adhering to all the requirements of the Revised Standardized Approach may not justify the incremental costs. Therefore, in keeping with the principle of balancing greater risk-sensitivity with reduced complexity, OSFI's view is that the added complexity of the Revised Standardized Approach is not appropriate for most SMSBs.

### 4.3 Proposed changes to the operational risk proposals

Given the feedback received and the noted deficiencies in the BIA, OSFI is proposing a simplified version of the Revised Standardized Approach ("Simplified Standardized Approach") as the default for all SMSBs. However, Medium-sized Institutions will have the ability to apply to use the full Revised Standardized Approach, if they can meet certain requirements.

OSFI's new proposed Simplified Standardized Approach is based on the existing BIA, but includes four adjustments to gross income, from the Revised Standardized Approach, that improve risk sensitivity by addressing some of the deficiencies in the BIA noted above, and are easy for institutions to report. The table below compares the BIA to the proposed Simplified Standardized Approach, with the four changes in bold:

Table 6
Current Basic Indicator Approach (BIA) Proposed Simplified Standardized Approach

Net Interest Income (including dividends)

+

+
Fee, commission and other income

=
Gross Income

x 15%

= Minimum Required Capital Operational Risk

Lesser of
Net Interest Income (including dividends), and
2.25% of Interest Earning Assets

+
Absolute value of Net Trading Income

+
Absolute value of P&L in the Banking Book

+
Fee, commission and other income

+
Share of income from Joint Ventures

=

x 15%

= Minimum Required Capital Operational Risk

Like the current BIA, all of the inputs in the calculation of the proposed Simplified Standardized Approach are from existing OSFI Income Statement (P3) and Balance Sheet (M4) returns. Also similar to the current BIA, the proposed Simplified Standardized Approach would use the 3-year average of each of the components when calculating adjusted gross income. Linked here is a template to calculate operational risk capital requirements using the Simplified Standardized Approach versus the BIA.

OSFI also recognizes that for some Medium-sized Institutions that have a history of collecting and using operational loss data for risk management purposes, the more risk-sensitive Revised Standardized Approach may be more appropriate. Larger, more complex institutions are expected to have operational risk management practices commensurate with their size and complexity, including the use of internal loss data. Therefore, OSFI is proposing that Medium-sized Institutions with annual adjusted gross income, based on the Simplified Standardized Approach (Adjusted Gross Income), greater than \$1.5 billionFootnote 13 would be required to use the Revised Standardized ApproachFootnote 14 with loss data.

Other Medium-sized Institutions (i.e., those with annual Adjusted Gross Income below \$1.5 billion) would also be able to apply to use the Revised Standardized Approach if they can demonstrateFootnote 15 that, in addition to prudent operational risk management practices, systems, and controls, as per OSFI's Guideline E-21: Operational Risk Management, they have:

• robust processes and procedures for the identification, collection, validation and use of operational loss data;
• processes to independently review the comprehensiveness and accuracy of loss data and ensure that they meet OSFI's expectations on data maintenanceFootnote 16;
• at least 10 years of quality historical loss data that meet the criteria detailed in the Basel III Operational Risk guidanceFootnote 17; and,
• the ability to map general ledger and/or relevant OSFI income statement returns to the line items required to calculate and report the Basel III "Business Indicator Component".

We also expect that upon receiving OSFI's approval to use the Revised Standardized Approach with loss data for capital calculation purposes, institutions would be required to file OSFI's L3 Return, Operational Risk Event Data Reporting Schedule, on a go-forward basis.

### 4.4 Operationalization of the use of Revised Standardized Approach

For the initial implementation of the Revised Standardized Approach in Q1 2022, the threshold calculation would be performed based on the Adjusted Gross Income for Medium-sized Institutions using the Simplified Standardized Approach for fiscal 2020. Then, starting in 2022, Adjusted Gross Income for Medium-sized Institutions would be calculated each year, at fiscal year-end. If Adjusted Gross Income is greater than \$1.5 billion, the institution would be required to use the Revised Standardized Approach in the following fiscal year. For example, if fiscal 2022 Adjusted Gross Income is greater than \$1.5 billion for the first time, the institution must use the Revised Standardized Approach starting in fiscal Q1 2024.

Institutions that reach the \$1.5 billion Adjusted Gross Income threshold, but do not have 10 years of loss data would be able to use 5-10 years of data, if this is available. Otherwise they would set the Internal Loss Multiplier equal to one in the Revised Standardized Approach.Footnote 18

Once the \$1.5 billion threshold is crossed, institutions must use the Revised Standardized Approach for a minimum of two years. If, after two years, Adjusted Gross Income falls below \$1.5 billion, the Medium-sized Institution may revert to the Simplified Standardized Approach.

### 4.5 Summary of operational risk proposals

The following table compares the operational risk treatment proposed in the July 2019 Discussion Paper to the treatment set out in this document.

Table 7
July 2019 Discussion Paper Proposal Jan 2020 Consultative Document Proposal
Category 1 & 2 Standardized Approach for Operational Risk Medium-sized Institutions
(Category I)
Simplified Standardized approach with option to apply to use the Revised Standardized Approach for Operational Risk (with loss data if requirements are met)
Category 3 Flat-rate capital add-on (TBD) Small Lenders
(Category II)
Simplified Standardized Approach
Category 4 No operational risk capital requirement Non-Lenders
(Category III)
Simplified Standardized Approach

OSFI believes that these revised proposals strike an appropriate balance between improving the risk-sensitivity and reducing complexity of operational risk capital requirements. The two main benefits are:

• Relatively comparable operational risk capital requirements within and across categories regardless of their business model or activities due to the consistent use of a revenue-based gross income indicator; and
• Improved risk sensitivity relative to the current approach used by most SMSBs (BIA), with minimal additional burden to institutions.

### 4.6 Questions on operational risk proposals

1. What are your views regarding moving from the current BIA to the proposed Simplified Standardized Approach?
2. What are your views on the proposed requirements that Medium-sized Institutions must meet in order to apply to use the Revised Standardized Approach (detailed in section 4.3)?
3. What are your views on the operationalization of the \$1.5B threshold, above which Medium-sized Institutions would be required to use the Revised Standardized Approach with loss data (detailed in section 4.4)?

## 5.0 Capital Requirements for Non-Lenders (Category III – formerly Category 4)

The July 2019 Discussion Paper proposed that institutions in Category 4 would no longer be subject to risk-based capital requirements given their smaller size and more limited complexity. Instead, they would be subject to a higher minimum leverage ratio, in the range of 8% to 12%, rather than the current minimum of 3%. The elimination of the risk-based capital requirements was intended to reduce the complexity of the capital requirements for these institutions while ensuring that they continue to hold adequate capital as determined by the leverage ratio.

### 5.1 Specific feedback received on the July 2019 Discussion Paper (Category 4 Capital requirements)

Many respondents were supportive of a simpler capital requirement for Category 4 institutions, such as the leverage ratio. However, a number of stakeholders were opposed to Category 4 institutions, particularly small lenders, having a higher minimum leverage ratio that could make it more challenging for them to compete with institutions in other categories, as well as provincially-regulated entities. Another stakeholder was concerned that a higher minimum leverage ratio could unfairly penalize new, but growing, institutions that are in Category 4 due to their small balance sheet size. Some respondents also suggested that small lenders would prefer to have more risk sensitivity in their capital requirements for competitive reasons. There was also a suggestion that capital requirements should be segmented by business activity, rather than just the size of the balance sheet.

### 5.2 Change to proposed capital requirement for Non-Lenders (Category III – formerly Category 4)

As described above in Section 2.2, OSFI is proposing changes to the segmentation criteria to ensure greater consistency of capital requirements for institutions with similar activities. All SMSBs with greater than \$100MM in loans, including newly established institutions that are expected to be above this threshold based on their business plan, should be in the Small Lenders category.

With the new proposed segmentation, the Non-Lender category will mostly be composed of institutions with non-lending business models. Operational risk is generally a more material source of risk than credit risk for these institutions, who either manage or administer off-balance sheet assets, or have other fee-based business activities. For these Non-Lenders, the leverage ratio, on its own, would not be a suitable capital requirement since it measures the size of an institution by the size of its balance sheet, and does not account for the operational risk of its activities.

OSFI considered retaining the current risk-based capital requirements instead of the leverage ratio as the only capital requirements for the Non-Lenders. However we do not believe this strikes the right balance between improving the risk sensitivity of the requirements for SMSBs and reducing the complexity of the frameworks to make them more fit for purpose. In particular, granular credit risk-weight reporting (through OSFI's BCAR) may be excessively complex and burdensome for some Non-Lenders where operational risk is generally a more material risk.

Therefore, OSFI is proposing a Simplified Risk-based Capital Ratio that replaces credit risk-weighted assets with total balance sheet assets in the ratio (see last column in Table 8 below).

Table 8
Current Leverage Ratio Current Risk-Based Ratios Proposed Risk-Based Ratio
$\frac{\mathrm{Tier 1 Capital}}{\mathrm{Total Assets \left(less deductions from capital and other adjustments for off-balance sheet exposures\right)}}$ $\frac{\mathrm{CET1/Tier 1/Total Capital}}{\mathrm{Credit RWA + Market RWA + Operational RWA}}$ $\frac{\mathrm{CET1 Capital}}{\mathrm{Total Assets \left(less deductions from capital\right) + Operational RWA}}$
3% minimum (with institution-specific Authorized levels) 7%/8.5%/10.5% minimum requirementsFootnote 19 10.5% minimum requirementFootnote 20

For the purposes of this ratio, capital will be Common Equity Tier 1 Capital (CET1). Generally, Non-Lenders do not have Additional Tier 1 or Tier 2 capital, therefore for simplicity OSFI is proposing to have only one CET1 minimum equal to the current total capital minimum requirement (i.e., 10.5%, which includes the capital conservation buffer)Footnote 21. This also ensures that the proposed minimum ratio would not reduce the stringency of the capital requirements for these institutions.

Total assets would be derived from the financial statementsFootnote 22, less any deductions from capital for intangibles and other adjustments (e.g., goodwill) as defined in Chapter 2 of OSFI's CAR Guideline. Operational risk-weighted assets would be calculated using the Simplified Standardized Approach described in Section 4 of this document.

This proposed capital measure would significantly reduce capital reporting requirements for Non-Lenders. Linked here is a template to calculate the proposed ratio that highlights all the data that would be needed (see the second tab labeled Simp. Risk-Based Capital Ratio).

### 5.3 Summary of capital requirement proposal for Non-Lenders (Category III – formerly Category 4)

The following table compares the Category 4 capital requirement proposed in the July 2019 Discussion Paper to the new proposed simplified risk-based capital ratio detailed above.

Table 9
July 2019 Discussion Paper Proposal Jan 2020 Consultative Document Proposal
Category 4

Leverage ratio
(set at 8-12% vs. current 3%)

Non-Lenders
(Category III)
New Simplified Risk-based Capital Ratio (minimum 10.5%):

$\frac{\mathrm{CET1 Capital}}{\mathrm{Total Assets + Operational RWA \left(using Simplified Standardized Approach\right)}}$

Under this proposal, the current leverage ratio and risk-based capital requirements for Non-Lenders would be replaced with a simpler, more streamlined, capital ratio that is more fit-for-purpose for these institutions. This proposed requirement achieves a balance by significantly reducing the complexity of credit risk requirements for Non-Lenders, while maintaining risk-sensitivity related to operational risks and the overall stringency of the current requirements.

### 5.4 Questions on capital requirement for Non-Lenders (Category III – formerly Category 4)

1. Do you believe the proposed new single requirement strikes the right balance between improving the risk sensitivity and reducing the complexity of Non-Lenders' capital requirements?
2. Do you agree with the proposal to have one capital requirement, based on CET1 capital, as opposed to the three current requirements for CET1, Tier 1 and Total capital?

## 6.0 Liquidity Risk

In the July 2019 Discussion Paper, OSFI proposed conceptual revisions to the liquidity framework for SMSBs covering the following main areas: i) application of the NSFR minimum standard to Category 1 (IRB) institutions only; ii) incorporation of a to-be-defined simplified liquidity metric for Category 3 institutions to complement the LCR minimum standard; and iii) replacement of the LCR and Net Cumulative Cash Flow (NCCF) metrics for Category 4 institutions by a to-be-determined simplified liquidity metric.

### 6.1 Specific feedback received in response to the July 2019 Discussion Paper (Liquidity Risk)

The majority of feedback received on the July 2019 Discussion Paper centred on three key themes: i) the scope of application of the NSFR minimum standard; ii) the utility of the NCCF for smaller institutions in the SMSB universe; and iii) the possible approaches that could be used in designing simpler liquidity metrics for Category 3 and/or Category 4 institutions. Each of these key themes is discussed in more detail below.

Regarding the scope of application of the NSFR minimum standard, several respondents suggested that the decision should be decoupled from whether or not an SMSB had received approval to use IRB for regulatory capital purposes. Instead, it was suggested that it be centered on risk-based principles that reflect SMSBs' typically simpler balance sheets and funding models. Respondents also noted that the proposed NSFR treatment may act as a disincentive for SMSBs to pursue funding diversification given the lower funding credit given in the metric to wholesale funding products compared to, for example, insured brokered deposits.

Regarding the application of simplified liquidity metrics for Category 3 institutions to complement the application of the LCR, several respondents supported the continued use of the NCCF as a tool to manage liquidity beyond the LCR's 30-day horizon. However, concerns were raised that NCCF reporting prescribes a level of detail that may be difficult and impractical for smaller institutions to operationalize. Specifically, those that cannot collect this level of data granularity due to systems or related process limitations/costs may be forced to make assumptions that may penalize the NCCF survival horizon output (e.g., identifying operational versus non-operational deposits may be a challenge due to data constraints). A suggestion was made that smaller institutions could provide a less detailed, more streamlined NCCF data filing to OSFI, as opposed to not producing the NCCF at all. Regarding calibration of the NCCF metric, one respondent suggested that OSFI revisit the prescribed run-offs as rates are not aligned with institution behaviour, in particular the treatment of unsecured wholesale term deposits. Another respondent suggested that off-balance sheet commitments be included as outflows. Finally, concern was expressed that if the proposed simplified liquidity approaches for Category 3 require these institutions to hold higher levels of liquidity than the current regime, this may have the unintended consequence of driving differentiated competitive advantages between institutions.

Feedback related to the proposed changes to liquidity requirements for Category 4 institutions revealed support for the proposed removal of the LCR as a minimum standard given respondents' views that the LCR does not consider assumptions that are relevant for non-lenders. Respondents noted that the alternative proposed metric focussed on a minimum amount of liquid assets as percentage of total assets would not be appropriate as it would be overly simplistic and would ignore variability in cash flows from month to month. Respondents generally thought a simple maturity mismatch or maturity gap report based on contractual cash outflows and cash inflows that does not consider the complexity of prescribed retention rates would be more appropriate. Some respondents also added that, given the uniqueness of business models for institutions in this category, such a measure should include operational costs and recognize deposits placed at other DTIs as a high-quality source of liquidity. Finally, respondents noted that any new liquidity measure should be carefully calibrated so that it does not impose significant increases in High Quality Liquid Asset (HQLA) holding requirements over current LCR and NCCF requirements.

Additional comments were noted on the frequency of regulatory reporting where some suggested that liquidity reporting requirements should be completed on a quarterly, not monthly, basis. Further, some recommended that the frequency of Liquidity Activity Monitor (LAM) reporting should be reduced as their view is that their institution's liquidity position does not fluctuate significantly on a weekly basis, leading some respondents to view LAM reporting as administratively burdensome.

### 6.2 Proposed Changes to the Liquidity Risk Proposals

#### Medium-sized Institutions (Category I)

Given the fundamental role the LCR and NCCF measures play in OSFI's liquidity adequacy framework, OSFI will require Medium-sized Institutions to continue to calculate the LCR and NCCF. Further, OSFI will look at revisions to the NCCF to improve the utility of this measure. As mentioned in the July 2019 Discussion Paper, this could take the form of modifications to increase risk capture (e.g., incorporate cash outflows related to commitments and operational expenses) and/or other modifications to existing cash flow assumptions (e.g., incorporate additional cash inflows related to residential mortgages, potential modifications to the run-off rates for unsecured wholesale term deposits). OSFI plans to consult in late spring 2020 on targeted amendments to the current NCCF assumptions that would be introduced in the Revised NCCF metric. Depending on the extent of refinements made to the Revised NCCF metric as part of that consultation and subsequent finalization, adjustments to any supervisory-communicated, institution-specific NCCF level that may currently be applied to an institution will be considered.

Regarding the NSFR, as noted in the LAR Guideline, the objective of this minimum standard is to reduce funding risk over a longer time horizon (than the 30-day LCR) by requiring institutions to fund their activities with sufficiently stable sources of funding. OSFI proposes to apply the NSFR to Medium-sized Institutions that significantly rely on wholesale funding sources. Specifically, Medium-sized Institutions that fund 40% or more of their total on-balance sheet assets with wholesale funding sources will be subject to the NSFR minimum requirement.Footnote 23

#### Small Lenders (Category II)

Small Lenders will continue to be subject to the LCR requirement as a base measure of liquidity that is comparable across institutions. While the LCR provides a perspective on cash flows and related liquid asset requirements over a 30-day time horizon, OSFI believes that it is important for institutions, and OSFI alike, to understand an institution's cash flows beyond the 30-day horizon in order to capture the risk posed by funding mismatches between assets and liabilities. The NCCF metric is a useful metric to identify gaps between contractual inflows and outflows for various time bands over 30 days and up to a 12-month time horizon, which indicate potential liquidity shortfalls an institution may need to address.

For this reason, OSFI will continue to require Small Lenders to be subject to a NCCF metric. However, OSFI believes that there is scope to significantly reduce the level of data required to be submitted by these institutions on a monthly basis. As such, OSFI proposes to create a Streamlined NCCF metric.

The assumptions in the Streamlined NCCF metric would largely be aligned with those included in the Revised NCCF metric described above to ensure consistency and comparability in outcomes across the two metrics.Footnote 24 However, OSFI will scale back the data granularity required of Small Lenders for NCCF reporting along the following areas:

• Reporting of data will be required on only one currency worksheet (i.e., the 'Combined NCCF' worksheet, which represents the CAD-equivalent of all currencies on a consolidated basis) in the Streamlined NCCF metric, rather than also on multiple currency worksheets (e.g., CAD, USD) as in the current NCCF template.
• Within the 'Assets' section, greater alignment will be incorporated in the Securities section with the categorization related to the breakdown of HQLA included in the LCR (e.g., Level 1, Level 2A, Level 2B) and non-HQLA securities.
• Fixed-term retail and small business (RSB) deposits that are assigned the same run-off rate will be collapsed within a single category, which will remove the current granularity related to the original term of the deposit (e.g., categories would be collapsed to totals for RSB Fixed Term Deposits – Type 1, insured, stable; for RSB Fixed Term Deposits – Type 1, insured, less stable; for RSB Fixed Term Deposits – Type 2, insured, stable; etc.).
• The detail currently reported under both the 'Repo and Securities Lent' and the 'Reverse Repo and Securities Borrowed' categories will be reduced to collect only information on the high-level sub-category breakdown in the current NCCF template (e.g. at the levels of total government securities, total MBS, total corporate bonds and paper, total ABS and ABCP, total equities and total "other").
• The collateral type breakdown under the 'Securities Sold Short' category will be removed such that only a total would be provided for the various sub-categories (e.g., total government securities, total MBS, total corporate bonds and paper, total ABS and ABCP, total equities and total "other").
• The ‘Collateral’ panel will be streamlined where identical haircuts allow for an aggregation of collateral types within each sub-category – e.g., total government securities, total MBS, total corporate bonds and paper, total ABS and ABCP, total equities and total “other” – and segments – e.g., pledging and encumbrances (derivatives and clearing), reverse repo collateral in, and repo collateral out.

A visual depiction of the structural revisions proposed to the current NCCF template that will create the Streamlined NCCF return is presented as a separate document (see link).

In addition, given their generally less complex funding structure, Small Lenders will not be subject to an NSFR requirement.

#### Non-Lenders (Category III)

In the July 2019 Discussion Paper, OSFI proposed that Category 4 institutions no longer be required to adhere to the LCR nor NCCF metrics going forward, nor would they be required to adhere to the NSFR. OSFI continues to hold this view.

Under the proposed segmentation changes related to Category 4 institutions mentioned in Section 2, the institutions that will be classified as Category III Non-Lenders do not primarily engage in financial intermediation. Given the nature of these institutions' activities, OSFI proposes to incorporate a relevant cash flow metric for Non-Lenders that is based on a simple maturity mismatch, including contractual outflows and inflows that factor in limited behavioural aspects captured by prescribed inflow and outflow rates.

As such, the Operating Cash Flow Statement (OCFS) metric will be applied to Non-Lenders. This metric would be structured to provide data on contractual cash inflows, contractual cash outflows and high-quality liquid assets over a one-year time horizon (i.e., weekly perspective for the first four weeks, monthly perspective for month 2 to month 12). Moreover, the type of information provided by institutions would include tailored categories that are relevant to the types of business activities conducted by Non-Lenders. For example, within the cash inflows section, data would be collected on categories such as fee-related inflows, interest on investments, interest on loans, and maturing loans. Similarly, within the cash outflows section, data would be collected on categories such as operational expenses, payroll-related outflows, maturing term deposits, demand deposits (where applicable), and interest payable. A full depiction of the proposed OCFS return template is provided at the following link.

In addition, OSFI is considering requiring individual institutions to meet a supervisory-communicated, institution-specific OCFS level. When determining the OCFS level for individual institutions, OSFI will consider such factors as the institution's operating and management experience, strength of parent, earnings, diversification of assets, type of assets, inherent risk of a business model and risk appetite. OCFS levels will be determined after consultation on the calibration of the OCFS metric, which is scheduled to occur in late spring 2020.

#### Summary

Table 10 includes a summary of the proposed changes to the SMSB liquidity requirements, compared to the July 2019 Discussion Paper proposals.

Table 10
July 2019 Discussion Paper proposal Jan 2020 Consultative Document proposal
Category 1 LCR; Revised NCCF; NSFR Medium-sized Institutions (Category I) LCR; Revised NCCF; NSFR only if significant reliance on wholesale funding (>40%)
Category 2 LCR; Revised NCCF
Category 3 LCR; simplified liquidity metric(s) Small Lenders (Category II) LCR; Streamlined NCCF
Category 4 Simplified liquidity metric(s) Non-Lenders (Category III) Operating Cash Flow Statement metric

OSFI believes that the proposed changes to the liquidity requirements provide the following benefits:

• The application of the NSFR minimum standard only to Medium-sized Institutions that have significant reliance on wholesale funding aligns directly with the objective of the NSFR to require institutions to fund their activities with sufficiently stable sources of funding.
• The combination of the application of the LCR and the Streamlined NCCF for Small Lenders will ensure that these institutions remain subject to adequate minimum liquidity requirements appropriate to their liquidity needs and generally less complex funding structure, while reducing the complexity of the framework for these institutions.
• Use of the Operating Cash Flow Statement metric for Non-Lenders better aligns with capturing the liquidity risk inherent in their unique business activities, and with the general lack of complexity of their structure and operations.

### 6.3 Operationalization of the liquidity requirements

The allocation of an institution to a given segmentation category, which will determine the liquidity requirements applied to the institution, would follow the operationalization of the segmentation criteria, described in Section 2.4.

For Medium-sized Institutions, the operationalization of the 40% wholesale funding reliance threshold for determining the scope of application of the NSFR minimum standard would be as follows:

• For the initial implementation in Q1 2022, the threshold calculation would be performed based on M4 data, as described in Appendix 2, covering the five fiscal quarter-end dates between fiscal Q2 2020 and Q2 2021.
• The application of the NSFR minimum requirement for periods after initial implementation in Q1 2022 would be determined based on a moving average covering M4 data from the previous five fiscal quarter-end dates, as described in Appendix 2. If a Medium-sized Institution exceeds the 40% threshold for the average of the previous five quarterly periods, the institution would be required to adhere to the 100% NSFR minimum standard as of the end of the fiscal quarter occurring nine months after the last quarterly reference date in the moving average period calculation Footnote 25. First reporting of the NSFR, using the template provided for regulatory reportingFootnote 26, would be expected to be provided to OSFI based on data as at that same fiscal quarter-end date.
• In cases where a Medium-sized Institution that is subject to the NSFR falls below the wholesale funding reliance threshold for a given five quarter moving average period, the institution would be required to continue to adhere to the NSFR minimum standard and report its NSFR position to OSFI until such time as it has demonstrated to OSFI that it is below the wholesale funding reliance threshold for four consecutive moving average periods.

### 6.4 Questions on Liquidity Risk proposals

1. Do you have any comments on the level or operationalization of the 40% wholesale funding threshold to determine the application of the NSFR for Medium-sized Institutions? Do you have any comments on the proposed calculation outlined in Appendix 2 that would determine application of the NSFR?
2. Regarding the items that are noted as possible amendments to the Revised NCCF and Streamlined NCCF metrics (i.e., cash outflows related to commitments, cash inflows related to residential mortgages, modifications to the run-off rates for unsecured wholesale term deposits), do you have specific comments on calibration levels that OSFI should consider?
3. In addition to the items mentioned in Question 2 that OSFI is already considering for the Revised NCCF and Streamlined NCCF metrics, do you have any further suggestions on revisions to the NCCF metrics' assumptions?
4. What are your views on the proposed structure of the Streamlined NCCF that would be applied to Small Lenders? Are there other items in the proposed Streamlined NCCF return that warrant consideration for additional streamlining (and if so, why)?
5. What are your views on the Operating Cash Flow Statement metric that would be applied to Non-Lenders? Are there other relevant cash inflow or cash outflow categories that should be incorporated as line items in the proposed OCFS return (and if so, why)?
6. What are your views on applying limited prescribed rates for select cash inflow/outflow categories when calculating the Operating Cash Flow Statement metric? Are there key cash inflow/outflow categories that prescribed haircuts/run-offs should be applied to?

## 7.0 Format/structure of future guidance

Following consultation on this document, OSFI plans to engage in consultation on more detailed frameworks for SMSBs. OSFI proposes that the SMSB capital and liquidity frameworks take the form of a guideline which outlines the segmentation of institutions and provides an overview of the applicable Pillar 1 requirements. The guideline would also include references to the relevant sections of more detailed guidelines on capital (CAR), liquidity (LAR) and leverage (LR).

This new guideline for SMSBs would bring all the requirements together in a single document, with references on where to look for more details, which should enhance the comprehensiveness of the framework. In addition, to improve the readability of the CAR Guideline, OSFI plans on incorporating more summary tables and decision trees (where appropriate), to help guide users through the requirements.

### 7.1 Questions on format/structure of future guidance

1. Do you have any comments on the proposed format/structure of the future guidance on SMSB capital and liquidity requirements?

## 8.0 Next Steps

In this consultative document, OSFI has proposed a number of changes to the Pillar 1 capital and liquidity requirements for SMSBs, and provided details regarding how the frameworks could be operationalized. Feedback received in response to this paper will inform the development of draft guidance which will be published in late spring 2020. OSFI is targeting publication of a final set of rules by December 2020 with an effective implementation date of Q1 2022.

The following table outlines the key milestones and timelines related to the implementation of the new SMSB capital and liquidity requirements:

Milestone Date
Publication of draft revisions to CAR, LAR, LR guidelines and new proposed SMSB capital and liquidity guideline Late spring 2020
Issuance of final SMSB capital and liquidity frameworks December 2020
Implementation of SMSB capital and liquidity frameworks Q1 2022

As well, in the late spring of 2020, OSFI will begin consultation with SMSBs regarding the content and frequency of regulatory reporting associated with Pillar 1 capital and liquidity requirements.

As noted in the Introduction, future phases of this initiative which relate to Pillar 2 and Pillar 3 capital and liquidity-related requirements will be the subject of consultations later this year.

Comments on this consultative document and responses to the questions included in the paper are requested by March 6, 2020 and should be sent to SMSB.Proportionality@osfi-bsif.gc.ca.

Please refer to Appendix 3 for a complete list of the questions included in this document.

## Appendix 1 – Summary Comparison of Proposals in July 2019 Discussion Paper and January 2020 Consultative Document

July 2019 Discussion Paper proposal
Capital Liquidity
Risk-based Capital Ratios Leverage Ratio (LR) Liquidity Coverage Ratio (LCR) Net Cumulative Cash Flow (NCCF) Net Stable Funding Ratio (NSFR)
Credit Risk Operational Risk

Category 1:

IRB approved SMSBs

Internal-Ratings Based Approach Standardized Approach LR LCR Revised NCCF NSFR

Category 2:

SMSBs >\$10B assets

Standardized Approach Standardized Approach LR LCR Revised NCCF No NSFR

Category 3:

SMSBs with
\$0.5B < Assets ≤ \$10B

or

>\$20B in AUM/AUA

Simplified Standardized Approach Flat rate capital add-on LR LCR No NCCF
Simplified liquidity metrics
No NSFR

Category 4:

SMSBs with
≤\$0.5B assets

and

≤\$20B in AUM/ AUA

No Risk-based Capital Ratios A higher minimum LR No LCR No NCCF
Simplified liquidity metrics
No NSFR
January 2020 Consultive Document proposal
Capital Liquidity
Risk-based Capital Ratios Leverage Ratio (LR) Liquidity Coverage Ratio (LCR) Net Cummulative Cash Flow (NCCF) Net Stable Funding Ratio (NSFR)
Credit Risk Operational Risk

Medium-sized Institutions
(Category I):

SMSBs > \$10B assets

Standardized Approach or
Internal-Ratings Based Approach

Simplfied Standardized Approach

Or

Standardized Approach with loss data

LR LCR Revised NCCF NSFR (based on extent and nature of wholesale funding reliance)

Small Lenders
(Category II):

SMSBs with Assets ≤\$10B

and

total loans >\$0.1B

Standardized Approach

(with simplified treatment available)

Simplified Standardized Approach

LR

LCR

Streamlined NCCF

No NSFR

Non-Lenders
(Category III):

SMSBs with ≤\$10B assets

and

≤\$0.1B in total loans

Simplified Risk-based Capital Ratio

No LCR

Operating Cash Flow Statement

No NSFR

## Appendix 2 – Calculation of Wholesale Funding Threshold for NSFR Application

The calculation of the wholesale funding reliance threshold for determination of the application of the NSFR minimum standard to Medium-sized institutions (Category I) is given by the following equation:

Where:

$\frac{\mathrm{Wholesale funding balances}}{\mathrm{Total on balance sheet assets}}≥40%$

Wholesale funding balances includes the sum of the following data point addresses (DPAs) found on OSFI's Balance Sheet (M4) regulatory return:

• Liabilities, demand and notice deposits:
• DPA 0873: Federal and provincial, total
• DPA 0874: Municipal or school corporations, total
• DPA 0875: Deposit-taking institutions, total
• DPA 0878: Other, total
• Liabilities, fixed-term deposits:
• DPA 0880: Federal and provincial, total
• DPA 0881: Municipal or school corporations, total
• DPA 2202: Deposit-taking institutions, total
• DPA 2339: Other than Canada/provinces/deposit-taking institutions/individuals, total
• DPA 2345: Liabilities, acceptances, total
• Liabilities, other liabilities:
• DPA 0620: Liabilities of subsidiaries other than deposits, call and other short loans payable, total
• DPA 0624: Liabilities of subsidiaries other than deposits, other than call and other short loans payable, total
• DPA 0632: Obligations related to borrowed securities, total
• DPA 0634: Obligations related to assets sold under repurchase agreements, Of which obligations are to the Bank of Canada or other organizations of the federal government, total

Total on-balance sheet assets is represented by DPA 1045 on OSFI's Balance Sheet (M4) regulatory return.

Each of 'wholesale funding balances' (numerator) and 'total on-balance sheet assets' (denominator) should be calculated as the sum of the quarter-end positions for the last five fiscal quarters, which will capture roughly a one year period.

## Appendix 3 – Complete list of questions included in the consultative document

### Questions on the segmentation proposals

1. Do you have any comments regarding the new proposed segmentation criteria including the criteria thresholds?
2. Do you have any comments regarding the operationalization of the segmentation process?

### Questions on credit risk proposals

1. Given the description of the credit risk standardized approach, do you think it is beneficial to have a simplified treatment available for Small Lenders to use for portfolios below the \$200 million threshold?
2. Do you have any views on the asset classes for which a simplified treatment is being proposed? Are there other asset classes for which a simplified treatment should be considered and, if so, why?
3. Do you have any comments on the level and/or operationalization of the \$200 million materiality threshold?

### Questions on operational risk proposals

1. What are your views regarding moving from the current BIA to the proposed Simplified Standardized Approach?
2. What are your views on the proposed requirements Medium-sized Institutions must meet in order to apply to use the Revised Standardized Approach (detailed in section 4.3)?
3. What are your views on the operationalization of the \$1.5B threshold, above which Medium-sized Institutions would be required to use the Revised Standardized Approach with loss data (detailed in section 4.4)?

### Questions on capital requirement for Non-Lenders (Category III – formerly Category 4)

1. Do you believe the proposed new single requirement strikes the right balance between improving the risk sensitivity and reducing the complexity of Non-Lenders' capital requirements?
2. Do you agree with the proposal to have one capital requirement, based on CET1 capital, as opposed to the three current requirements for CET1, Tier 1 and Total capital?

### Questions on Liquidity Risk proposals

1. Do you have any comments on the level or operationalization of the 40% wholesale funding threshold to determine the application of the NSFR for Medium-sized institutions? Do you have any comments on the proposed calculation outlined in Appendix 2 that would determine application of the NSFR?
2. Regarding the items that are noted as possible amendments to the Revised NCCF and Streamlined NCCF metrics (i.e., cash outflows related to commitments, cash inflows related to residential mortgages, modifications to the run-off rates for unsecured wholesale term deposits), do you have specific comments on calibration levels that OSFI should consider?
3. In addition to the items mentioned in Question 2 that OSFI is already considering for the Revised NCCF and Streamlined NCCF metrics, do you have any further suggestions on revisions to the NCCF metrics' assumptions?
4. What are your views on the proposed structure of the Streamlined NCCF that would be applied to Small Lenders? Are there other items in the proposed Streamlined NCCF return that warrant consideration for additional streamlining (and if so, why)?
5. What are your views on the Operating Cash Flow Statement metric that would be applied to Non-Lenders? Are there other relevant cash inflow or cash outflow categories that should be incorporated as line items in the proposed OCFS return (and if so, why)?
6. What are your views on applying limited prescribed rates for select cash inflow/outflow categories when calculating the Operating Cash Flow Statement metric? Are there key cash inflow/outflow categories that prescribed haircuts/run-offs should be applied to?

### Questions on format/structure of future guidance

1. Do you have any comments on the proposed format/structure of the future guidance on SMSB capital and liquidity requirements?

## Footnotes

Footnote 1

OSFI Discussion Paper Advancing Proportionality: Tailoring Capital and Liquidity Requirements for Small and Medium-Sized Deposit-Taking Institutions, July 2019

Footnote 2

Pillar 1 requirements include the Capital Adequacy Requirements (CAR) Guideline, the Leverage Requirements (LR) Guideline and the Liquidity Adequacy Requirements (LAR) Guideline.

Footnote 3

Pillar 2 requirements refer to OSFI's guidance that outlines prudential and risk management expectations related to capital and liquidity as well as associated supervisory oversight. Pillar 3 requirements refer to OSFI's guidance on public disclosure requirements related to capital and liquidity.

Footnote 4

The initial categorization for new federally regulated institutions will be based on the business plan provided during the application process.

Footnote 5

Total assets would be calculated from line 7 of the M4 return. Total loans would be calculated from line 3 of the M4 return.

Footnote 6

The 2019 CAR Guideline includes a more complete description of what should be included in a particular asset class. However, these definitions are subject to revision as the domestic implementation of the Basel III reforms are finalized. OSFI expects to publish draft revisions to CAR Chapter 3 (Credit Risk – Standardized Approach) for consultation in late spring 2020.

Footnote 7

Excluding commercial real estate as defined in footnote 8.

Footnote 8

A commercial real estate exposure is an exposure secured by any immovable property that is not residential real estate. (Residential real estate is defined as an immovable property that has the nature of a dwelling and satisfies all applicable laws and regulations enabling the property to be occupied for housing purposes, such as individual condominium residences and one-to four-unit residences).

Footnote 9

Basel Committee on Banking Supervision: Basel III: Finalising post-crisis reforms, December 2017 (https://www.bis.org/bcbs/publ/d424.pdf)

Footnote 10

Total exposures includes both on and off balance sheet, net of Stage 3 allowances but before taking into account credit risk mitigation.

Footnote 11

Please refer to the BCBS document "Operational Risk: Revisions to the Simpler Approaches", October 2014, which describes in more detail the rationale for the revision of the BIA proxy indicator for operational risk.

Footnote 12

For example, under the Revised Standardized Approach, fee and commission income and expenses must be reported on a gross basis, and institutions must include either three or ten years of operational loss data.

Footnote 13

The \$1.5 billion threshold is aligned with the BCBS framework on operational risk capital, which requires banks with a Business Indicator Component greater than €1 billion to use loss data in the Revised Standardized approach (see link in footnote below).

Footnote 14

OSFI expects to publish draft revisions to CAR Chapter 8 (Operational Risk) for consultation in late spring 2020. The Revised Standardized approach (Basel III) will be substantively based on the guidance posted by the BCBS in December 2017 (https://www.bis.org/bcbs/publ/d424.pdf).

Footnote 15

This may include a self-assessment with an independent validation (e.g., from the institution's internal audit function).

Footnote 16

OSFI's expectations on data maintenance related to the Revised Standardized Approach are still to be finalized. OSFI expects to consult on these expectations as part of the draft revisions to CAR Chapter 8 (Operational Risk) in late spring 2020. OSFI will consider the current expectations for "Data Maintenance at TSA and AMA Institutions" as these revised expectations are developed (see link to OSFI implementation note here https://www.osfi-bsif.gc.ca/eng/docs/op_risk_datamaint.pdf).

Footnote 17

See Section 5 (page 130-131) of the Basel Committee on Banking Supervision document: Basel III: Finalising post-crisis reforms, December 2017 https://www.bis.org/bcbs/publ/d424.pdf. These requirements will be incorporated into OSFI's CAR Chapter 8 (Operational Risk) for implementation in 2022.

Footnote 18

OSFI expects to publish draft revisions to CAR Chapter 8 (Operational Risk – Standardized Approach) for consultation in late spring 2020.

Footnote 19

Note that these requirements include the 2.5% capital conservation buffer, as per section 1.6.1 of OSFI's CAR Guideline.

Footnote 20

This requirement would include the 2.5% capital conservation buffer.

Footnote 21

Non-Lenders that issue Additional Tier 1 or Tier 2 capital would continue to have minimum CET1/Tier 1/Total Capital ratios of 7%/8.5%/10.5% respectively (these include the capital conservation buffer). This ensures institutions maintain a certain level of the highest quality capital (CET1) as part of their minimum regulatory capital requirements.

Footnote 22

Total assets would be calculated from line 7 of OSFI's M4 return.

Footnote 23

Specific calculation details regarding the threshold, including data points from existing OSFI returns used in its calculation, are provided in Appendix 2.

Footnote 24

Note that any targeted adjustments that will be incorporated in the Revised NCCF metric (as described above for Medium-sized Institutions) after consultation in late spring 2020 will also be incorporated in the Streamlined NCCF metric. In addition, depending on the extent of refinements made to the Streamlined NCCF metric as part of that consultation and subsequent finalization, adjustments to any supervisory-communicated, institution-specific NCCF level that may currently be applied to an institution will be considered.

Footnote 25

For example, assume that based on data submitted for the fiscal quarter ends from Q2 2022 to Q3 2023, an institution (with an October fiscal year end) exceeded the 40% threshold. In this case, the institution would be expected to adhere to the NSFR minimum requirement by the end of Q2 2024. First regulatory reporting of the NSFR to OSFI would be expected to occur based on end-April 2024 (fiscal Q2 2024) data.

Footnote 26

See NSFR regulatory return at https://www.osfi-bsif.gc.ca/Eng/fi-if/rtn-rlv/fr-rf/dti-id/Pages/NSFR_ccl.aspx.

​​​ ​