On June 8, 2021, S&P Global Ratings published a request for comment (RFC) on its proposed criteria, "Request For Comment: Banking Industry Country Risk Assessment Methodology And Assumptions."

Following feedback from market participants, we finalized and published our criteria, "Banking Industry Country Risk Assessment Methodology And Assumptions" (BICRA criteria), on Dec. 9, 2021.

We'd like to thank investors, issuers, and other intermediaries who provided feedback. This RFC Process Summary provides an overview of the written comments and certain other feedback we received from the market on the proposed criteria, the changes we made following the RFC period, and the rationale for those changes. Several of the comments discussed in this article were submitted as part of a response to "Request For Comment: Financial Institutions Rating Methodology," June 8, 2021, where that response addressed items in both RFCs. We address all comments relating to the BICRA proposals in this article. See "RFC Process Summary: Financial Institutions Rating Methodology," Dec. 9, 2021, for a discussion of the comments received on the proposals in the other RFC.

In the RFC, we asked several questions:

• What are your views on the methodology we have discussed in this article?
• Are there any other factors you believe we should consider in the proposed criteria?
• In your opinion, do the proposed criteria contain any significant redundancies or omissions?
• Is the structure of the methodology clear, and if not, why?
• Do you believe this framework places too much emphasis on any particular credit factor, and if so, do you believe this emphasis could be mitigated by the use of the adjustments?
• Do you believe we are appropriately capturing risk and agree with the manner in which we propose to assess this risk? If not, what alternative(s) would you propose?

Respondents addressed these questions and other issues. Several respondents expressed a general appreciation for the transparency of the RFC process, and for the relevancy of the proposed changes to the BICRA criteria to address changing economic landscapes and industry characteristics, as well as to capture emerging risks and trends, including environmental, social, and governance (ESG) factors.

Written Comments Received From Market Participants That Led To Significant Analytical Changes To The Final Criteria

We did not receive any external written comments from market participants that led to significant analytical changes to the criteria.

Written Comments Received From Market Participants That Did Not Lead To Significant Analytical Changes To The Final Criteria

The forward-looking nature of a BICRA

We received some comments about the degree to which our BICRA assessment looks at future expectations or is affected by past events in a country. We confirm that the BICRA is a forward-looking assessment. We assign a BICRA score based on a forward-looking time horizon of three to five years, but also incorporate characteristics beyond this horizon where we consider them to be relevant, material, and sufficiently visible. Past events can help inform our expectations but do not limit the BICRA score or component factor scores in an automatic or mechanistic manner. For example, one comment remarked that our institutional framework score should not be held back due to past regulatory weaknesses if they have since been addressed. Another comment highlighted that the degree to which a banking system received state aid in past crises is useful information. We agree with both these views. Our scoring is forward looking but considers how a system has responded to or dealt with past events as we establish our view of how these events may or may not be relevant to the system's future path.

Impact of moving to a single adjustment to the initial score for each factor

Several comments addressed the potential impact of moving to a single adjustment to the initial score as opposed to the previous approach of having several individual adjustments each with a set impact on the final score. These respondents raised concerns that this could lead to greater subjectivity and make the reasons for adjustments less transparent. We agree that the reasons for making an adjustment should be transparent so the key reasons driving our scores will be discussed clearly in BICRA reports.

The new criteria do provide the potential for more flexibility in arriving at the adjustment than under the previous criteria but we believe that this will reduce the possibility of prescribed adjustments leading to a score that does not reflect the risks appropriately. This is because the single adjustment considers the holistic impact of various factors by taking account of the interplay between the various drivers of the adjustment. We expect very limited impact on actual outcomes due to this change from the previous criteria. We also believe that the single adjustment can help communicate the rationale more clearly by focusing on the characteristics that drive the adjustment and not on characteristics that are less relevant to the country whose banking system we are assessing. We will continue to support analytical consistency through our committee and analytical oversight processes.

Another comment raised a concern that the proposed approach would lead to greater use of negative adjustments and a skewing toward higher risk outcomes, linked to a view that there isn't enough credit given for characteristics such as macroeconomic policy flexibility, governance and transparency, and industry stability. We agree with the importance of reflecting strengths in each BICRA factor. We expect that the initial scores will now be better able to reflect such strengths, due for example to the capacity to use different metrics or indicators that highlight the specifics of that banking system. The BICRA framework does allow for negative adjustments to be larger in size than positive adjustments because this reflects our view, and experience from applying the previous criteria since 2011, that the potential impact of significant risks could be greater than the impact of mitigating factors. This does not mean that we will always apply negative adjustments more frequently than positive adjustments, and there are no limitations on how frequently we may apply positive adjustments.

Ability of BICRA to reflect regional features, characteristics relevant to a part of the banking system, or different types of structures

We received support for the increased ability to reflect different banking system structures through the ability to use different metrics or indicators. However, we also received several comments on how we should focus more on in-country regional differences in economic or industry risk, for example if there are concentrations to a specific region, and ensure that a bank's stand-alone credit profile (SACP) addresses this more clearly. We agree that in-country differences in industry risk and economic risk can be very relevant to the creditworthiness of a bank. We address this in "Financial Institutions Rating Methodology," Dec. 9, 2021 (our FI criteria) through the entity-specific assessments used to arrive at the SACP and rating on the bank. We assign BICRAs at the level of a country and do not have regional BICRA scores within a country except for very specific cases such as Hong Kong (where that territory has separate legal, regulatory, and reporting frameworks to other parts of China). Legal and regulatory systems tend to have similarities within a country, even if there can be regional differences, and the markets for banking products tend to have fewer barriers within a country than across national borders.

We recognize that in-country differences can lead to banks in one part of a country operating in a more attractive environment than those in other parts. When the countrywide economic risk or industry risk score does not represent the risks for that bank, we use the entity-specific scores in our FI criteria to reflect this. In our new FI criteria, we have for example increased the granularity of the scoring for several factors versus the previous criteria to address this and other entity-specific differences. We have also made several edits to the descriptors for the different categories that enhance the capacity to reflect in-country differences.

Business position and risk position are both scored on a relative, rather than an absolute basis. For example, a bank may have a good franchise in a part of a country but be assigned a weaker business position score to reflect the less attractive aspects of that part as a place in which to do business, which means that the countrywide industry risk score underrepresents the challenges of the market for that bank. If the region is a more attractive operating environment than the country, this could boost the business position score for a bank if the industry risk score sufficiently overestimates the challenges of the market for that bank. Similarly, a bank could have a weaker business position score if the industry risk score is boosted by regulatory approaches that do not apply to that type of bank. In the same way, a bank will have a weaker risk position score if the economic risk score for that country underrepresents the risks present in that part of the country (and vice versa). The new FI criteria also do not limit the frequency with which a bank can be assigned one of the higher or weaker scores for a category (for example the frequency with which banks can be scored as strong or very strong).

Another comment asked if a few weak banks could influence the outcome for all banks in a country and therefore lead to frequent changes in the BICRA. This could happen if the overall impact of those few banks on the system is strong enough but would not do so if their impact on the system is low. Even if the BICRA does change because of problems in a sector within the banking system, the rating on a specific bank would not change if we consider that its creditworthiness has not been affected, for example we could revise up the business position or risk position score for that bank if we believe that the new industry or economic risk scores overestimate the risk for that bank.

Comments related to economic risk

The impact of geographic features or a high reliance on agriculture on the economic resilience score.  One comment noted that high reliance on agriculture is not necessarily a cause of vulnerability and shouldn't automatically drive a negative adjustment to the economic resilience score. Another comment raised similar concerns about pulling down the economic resilience score due to geographic exposure to environmental risks without taking account of the capacity to mitigate such risk. We agree with such comments and have edited this section to clarify that we make a negative adjustment here when there are significant vulnerabilities to climate transition or physical risks, which does allow us to take account of the capacity to manage such risks. We have also edited the paragraph to show that the agricultural sector could be an example of where such vulnerabilities arise.

The role of a debt-to-GDP ratio as an indicator of financial crises.  We received two similar comments that focused on the limitations of using a debt-to-GDP ratio as an indicator of financial crises. Both comments agreed that the BICRA provides a framework to look at the cyclicality of a banking system and that expansion and correction phases are useful concepts to look at this. They suggested focusing on private sector credit growth and gross savings, as well as nominal GDP, instead of a debt-to-GDP ratio. We reviewed this comment in the context of two factors within our economic risk assessment: the economic imbalances score and the credit risk in the economy score.

We agree that private sector credit growth is more insightful than the level of debt when looking at economic imbalances. We therefore set our initial economic imbalances score using the change in private sector debt and not the absolute level. We also agree with the comment highlighting how changes in gross savings help identify rising or falling structural risks. Although savings rates could already influence the final economic imbalances score in the RFC proposals, we have added text to the section on the adjustments to the initial economic imbalances score to flag more prominently how reductions in savings rates can lead to a negative adjustment. Savings considerations can also result in a positive adjustment when that leads to the household sector being significantly stronger than suggested by the initial score.

These comments focused on how the level of household debt per se does not explain the reduction in household demand in a correction. For example, demand for purchases related to household lifecycle events can lead to a build-up of debt to finance those events. The saturation of that demand is relevant to what the change in private sector debt may mean for the economy. We agree with the commenter that a consideration of the drivers of credit growth (building on more granular household and cohort data as possible) and of the level and composition of household net wealth are important when looking at vulnerability of the banking system to an economic correction. We reflect this by adjusting the initial economic imbalances score when these considerations mean that the household or corporate sectors are stronger or weaker than the initial score suggests. Another commenter flagged that wealthier households can still experience serious loan delinquencies in a market correction. We agree with this observation and see this as highlighting the importance of looking at the patterns and drivers of the credit growth in that specific economy and not scoring based on headline metrics only.

Other characteristics suggested as indicators of overheating risk included GDP gap analysis and various sentiment indicators, citing how households have turned decidedly more negative in their view of the future about 12 months prior to the outbreak of a systemic financial crisis. We agree that these considerations are often very informative (even though the data can vary by timeliness, frequency, and definition or construction on a global basis) and expect such analysis to inform adjustments to the initial economic imbalances score by providing insight into the situation of a specific economy in that particular economic cycle. For example, measures of consumer and business sentiment inform our assessment of where the economy is within its economic cycle.

Another suggestion was that systemwide impairments rarely occur outside of systemic financial crises, and that therefore we should focus our analysis on identifying imbalances rather than assuming a normal rate of loss for such an inherently cyclical industry as banking. We agree that the economic imbalances score should focus on identifying the risks associated with imbalances. We consider it to be useful for our analysis to compare observed and expected credit losses and nonperforming loans (NPLs) in a system with the observed cyclical low points for that system when we are assessing where the system might be in that economic cycle. We do not score mechanistically based on these metrics, however.

That same comment suggested focusing on economic imbalances within our BICRA analysis. We agree that the analysis of economic imbalances is an important driver of economic risk for a banking system, particularly focusing on cyclical aspects that drive crises, but we consider that structural factors are also relevant because they have longer-term effects on the vulnerability to a systemic crisis. Our overall economic risk score therefore continues to be driven by not only economic imbalances but also the other two economic risk factors: economic resilience and credit risk in the economy. These factors may evolve more slowly than the economic cycle but we consider that they affect the way in which a banking system will react to cycles over time. For example, a system with weaker economic resilience may find it more difficult to deal with a correction phase.

The comment also suggested that one-year trends in some metrics can be more useful than a three-year period. (In our criteria, we typically look at the past three years plus the current year estimates, although we may use data for shorter or longer periods.) We consider that looking only at the trend over one year is often too short, for example because the data may only become available after a time lag or may be affected by shorter-term noise. In many countries, we see the build-up of imbalances over a period of three or more years before a correction and we often see fluctuations during the build-up phase rather than a steady trend. We agree with the comment that sharp changes over a one-year period can give important insight in various cases, however, and therefore we may assign the initial economic imbalances score by using data for a shorter period where relevant. We may also apply an adjustment to the initial score if we see something material and relevant occurring within a short timeframe. We do not expect to wait until we think there is only a year before a correction to lower the economic imbalances score, however, because we are not assessing the probability that banks will default over the next year or any other specific time period.

We do use private sector debt-to-GDP ratios to assess the credit risk in the economy factor (another factor within economic risk) because we consider the level of leverage in the economy to be an important factor within the overall economic risk. We compare various systems using this metric for our initial score, then adjust the score with more detailed information for that economy. Our adjustment to the initial score for this factor considers different characteristics relevant to that system. For example, the household sector's debt to GDP, debt service to disposable income, financial wealth, and delinquencies, as well as NPLs in the system. The adjustment also takes account of patterns within the household sector, for example where the overall risk may be over- or underestimated by aggregate data but driven more by the experience of certain cohorts. We agree with the comments that such granular analysis gives additional insight into the capacity of an economy to bear a particular degree of leverage and therefore reflect them where relevant in the adjustment to arrive at the final score.

Adjustments to the initial credit risk in the economy score.  One comment asked why we did not propose to allow a positive adjustment of two notches if the initial credit risk in the economy score is '3', given that a positive two-notch adjustment is possible at weaker initial scores. We have decided to maintain our proposed approach. We don't consider that a banking system with the degree of leverage commensurate with an initial score of '3' should receive the lowest-risk final score of '1' for credit risk in the economy. We can make a positive adjustment of one notch, however.

Comments related to industry risk

Institutional framework.  One comment asked that our analysis recognize the concept of regulatory proportionality, such that some regulators may not need to focus on certain risks or have developed certain sophisticated approaches due to the structure of their sector. We agree and for this reason address this, both in our proposals and final criteria, in our section on complexity (which can drive a positive adjustment to our institutional framework initial score) where we state that the regulation and supervision of a very simple banking sector would not require the same level of sophistication. In line with another comment, we also confirm that, in the case of a complex banking system, we assess whether regulators are well equipped to deal with this level of complexity. We do compare regulatory approaches with international standards, however, as part of our initial score (table 13 uses a "best fit" analysis to assign the initial score).

We also added text to clarify why we do not make a positive adjustment to the initial institutional framework score for governance and transparency reasons, in response to a comment. The initial score takes account of regulatory monitoring of ESG-related factors, which includes encouragement of governance and transparency initiatives.

We also received a comment that the existence of directed lending should not automatically lead to a negative adjustment for governance and transparency purposes because this lending could be correcting a market distortion. We continue to cite directed lending as an example of what may be negative intervention by a government that could affect the balance of stakeholder interests. Whether we make a negative adjustment will depend on the impact of such lending on the banking system.

Industry stability and new entrants.  One comment discussed how new entrants can improve the quality of a banking system and that the possibility of new entrants should not automatically lead to a more negative view of industry stability within our competitive dynamics score. We agree that the impact of new entrants can take various forms. We first consider what the potential impact of new entrants on industry stability would be. If we then conclude that it will lead to industry instability (for example if we expect that it will result in inadequate pricing for risk across the system), then this could lead to a negative adjustment, but not otherwise. In response to the comment, we made an edit to the criteria to make it clearer that we only make a negative adjustment to the initial score if we consider that the system faces sufficient additional risks due to instability.

Significant Analytical Changes To The Final Criteria That Did Not Arise From Market Feedback

We finalized and published the criteria without making any significant analytical changes that are unrelated to the external written comments or other market feedback we received.

However, we incorporated other editorial revisions, where appropriate, to improve the readability and clarify the intent of the criteria. For example, we removed duplicate text. We also aligned our wording of ESG-related factors, such as climate transition and physical risks, with the wording used in our other ESG-related publications. These revisions are not significant analytical changes.

In table 11, we removed a bullet point that gave an example of when we may apply a positive adjustment to the initial credit risk in the economy score. This was because that bullet point referred to cases where household debt is held predominantly by wealthier households, which can already be addressed by the other text in the table (such as the bullet point referring to when GDP per capita understates the debt capacity of the part of the private sector that borrows from banks). The deletion doesn't alter the analytical approach.

We also deleted the reference to using average ratios over the past two completed years in order to determine the initial systemwide funding score. This is because we typically base the scoring on our expectations for the likely trend in the near future. Our assessment is forward looking so we believe it is more accurate to state that our view is informed by the ratios for the past two completed years than that a specific average over that time period drives our view. This is not a significant analytical change.

We also provided the following clarifications:

• We clarified the situations when we use table 8 instead of table 7 to determine the initial economic imbalances score in an expansionary phase.
• We clarified why we might exclude certain items from our measure of private sector debt within an economy, so that the rationale for any adjustments would be clearer.
• We made several edits to table 14 to clarify that the initial score for competitive dynamics is based on our view of the banking sector's risk appetite.
• We also clarified when we would adjust the sovereign net external debt figure that is used in our sovereign rating analysis to arrive at the net external debt measure that we use for our initial systemwide funding score.
• To enhance consistency, we clarified the composition of the various components of core customer deposits, and that each of the components of "domestic" core customer deposits are from resident depositors.

Related Criteria

• Banking Industry Country Risk Assessment Methodology And Assumptions, Dec. 9, 2021
• Financial Institutions Rating Methodology, Dec. 9, 2021