North American CRO Council Issues Public Comment on Treasury Department Notice

WASHINGTON, Jan. 22 -- Jonathan Porter, chair of the North American CRO Council, has issued a public comment on the Department of the Treasury notice entitled "Federal Insurance Office Study on the Insurance Capital Standard". The comment was written on Jan. 15, 2021, and posted on Jan. 21, 2021:

* * *

The North American CRO Council (CRO Council) is a professional association of Chief Risk Officers (CROs) from leading insurers based in the United States, Canada, and Bermuda. Member CROs currently represent 32 of the largest Life and Property and Casualty (P&C) insurers in North America. The CRO Council seeks to develop and promote leading practices in risk management throughout the insurance industry and provide thought leadership and direction on the advancement of risk-based solvency and liquidity assessments.

The CRO Council appreciates the opportunity to comment on the request for information (RFI) issued by the Federal Insurance Office (FIO) to solicit input on FIO's planned study of the potential effects of the insurance capital standard (ICS) on US insurance markets, consumers, and insurers. We support FIO's ongoing engagement at the International Association of Insurance Supervisors (IAIS) and its continued collaboration with the National Association of Insurance Commissioners (NAIC) and Federal Reserve Board (Federal Reserve) to promote an international standard-setting process that reflects and respects the well-established enterprise risk management practices, regulatory capital regime, and supervisory oversight of the US insurance sector.

In this vein, among the various IAIS initiatives to be performed during the ICS Monitoring Period, we believe that FIO's top priority should be to support a meaningful and viable process for assessing the Aggregation Method (AM) as providing comparable outcomes to the ICS. FIO's proposed economic study should support this goal by demonstrating, in an evidence-based and objective manner, the shortcomings and potential unintended consequences that would arise if US regulators were to implement the ICS as a group regulatory capital requirement. In short, the comparability determination should reflect both the appropriateness of the Aggregation Method and, respectively, the significant flaws of the ICS as a group capital construct, which inappropriately treats capital owned by individual insurance entities as fully fungible.

We also urge FIO to rely as much as feasible on existing or ongoing research, analysis, and data collection efforts. For example, FIO should seek to leverage prior ICS Field Test data (subject to steps to ensure adequate confidentiality protections are in place); a forthcoming study by the Federal Reserve's Insurance Policy Advisory Committee (IPAC) to assess potential implications of the ICS for the US insurance industry, markets, and policyholders, with a specific focus on long duration life and annuity products; and prior impact analysis conducted by a global coalition of CROs, which was based on stylized but representative insurer portfolios (n.b., this CRO study was non-public, but its conclusions and modes of analysis could be useful in informing FIO's assessment of ICS design and methodology issues). We believe that, coupled with publicly-available data, the FIO would have ample information for conducting a substantive study and avoid the need to impose additional data requests on the industry beyond ongoing supervisory reporting and information gathering exercises.

Substantively, the FIO study should highlight the significant conceptual and technical shortcomings of the ICS. Notably, the market-adjusted valuation (MAV) basis for the ICS is both unproven as the basis for a global group regulatory capital standard and likely distortive in its current technical design. Market-based approaches can provide information value as an alternative lens for internal management reporting; however, they have shortcomings. If not carefully considered, these shortcomings can create unintended consequences and inappropriate risk management incentives. We believe the IAIS has yet to sufficiently address these shortcomings, which are particularly impactful for longer-dated insurance liabilities subject to numerous hardwired regulatory parameters that appear to conflict with asset and liability management (ALM) disciplines and economic realities. We believe the IPAC study will illuminate the technical flaws underlying the MAV approach, particularly for long-duration products. We therefore encourage FIO to coordinate its work with the IPAC study. The FIO study should primarily focus any original research and analysis on issues and concerns that have not been addressed fully in other contexts, namely:

* The ICS treatment of credit risk, notably the relevance of NAIC designations as economically-sensitive measures of fundamental credit risk, as well as the inappropriateness of non-default spread risk (NDSR) charges that asymmetrically impact US insurance groups;

* the role of senior debt and capital instruments in providing a reliable source of loss absorbing capital at the operating entity level to protect policyholders, particularly during periods of market stress when equity issuance is challenged;

* excess conservatism in risk calibrations, beyond the demonstrable risk exposure or loss experience during stress periods (e.g., mass lapse; market risks); and

* the significant costs of ICS implementation, not only in terms of operationalizing the MAV within a fully-controlled and audited environment, but also the large-scale and sustained outreach that would be required to educate stakeholders (i.e., jurisdictional supervisors, investors, rating agencies) on the technical aspects, drivers, and shortcomings of the resulting capital ratios.

Below are the CRO Council's responses to the individual questions posed in the RFI.

1. If the ICS were adopted in the United States, how would this affect the insurance market in the United States, including consumers and insurers? How would the adoption of the ICS affect the competitiveness of U.S.-domiciled IAIGs, foreign insurance groups with significant operations in the United States, and U.S. insurers that have current or planned operations abroad?

ICS adoption in the US would impose multiple adverse impacts on the cost of insurance coverage for consumers, as well as on well-established risk management and ALM practices within the insurance sector as a whole. The CRO Council expects that ICS adoption would require substantial implementation costs for both Life and Retirement (L&R) and Property and Casualty (P&C) insurers, with little discernible prudential utility.

Within L&R, increased costs due to ICS could, in turn, undermine the availability of long-duration products, thereby exacerbating the protection gap for retirement security and dis-incentivizing appropriate exposure to long duration assets. While it is important that a construct such as the ICS reflect secular trends in insurance risks and market conditions (such as the current "low for longer" interest rate scenario), the design and criteria underlying the ICS do not sufficiently recognize ALM conventions and asset performance. For example, the ICS synthetically constrains both the eligibility of assets and the amount of spread recognized in discounting the corresponding liabilities, which would result in sub-optimal ALM that puts policyholders at risk (e.g., by incentivizing investment strategies that do not generate sufficient income to defease insurance obligations). Particularly during the current low interest rate environment, it is important for capital frameworks, including ICS, to appropriately recognize that insurers back their long-term liabilities with a diverse set of long-term assets. To be clear, we are not advocating excess risk-taking to maximize shareholder returns, but rather the continued application of prudent, diversified ALM risk mitigation strategies to ensure that long-term returns will be sufficient to discharge long-term liabilities.

Investing in long-term real assets reduces solvency risk, and a framework that prevents or penalizes insurers for doing so could increase the likelihood of insurer insolvency over the long-term and, in parallel, reduce the availability of long-duration life insurance products. Moreover, if the ICS were to incentivize holding only those assets deemed "safe" by the IAIS (e.g., high-grade Corporates, US Treasuries, etc.), the resulting distortions in ALM would not only undermine policyholder protection, but could also adversely impact financial stability by driving insurers to "herd" into a narrow universe of eligible assets.

For P&C insurers, ICS adoption would entail sizable implementation costs for a wholly new construct that provides minimal informational value for short-duration business. Essentially, application of MAV for most P&C lines would result in some uptick in reported capital, by restating undiscounted US GAAP liabilities on a present value basis. However, in rough measure, this impact would be backfilled through the introduction of an explicit ICS risk margin (margin over current estimate, or MOCE). The application of ICS for P&C carriers would entail substantial mechanical efforts to generate broadly similar balance sheet results.

US implementation would entail significant costs not only in terms of valuation and reporting systems, but also in educating external stakeholders about the technical design and behavior of the ICS. We expect there to be potentially high transition or frictional costs in introducing an entirely new group capital regime like the ICS - a construct which US insurers do not currently manage to, are not regulated by, and whose properties and behavior are largely untested across the economic cycle. The costs and challenges would also apply to states and their respective insurance regulators. US state supervisors have largely not been involved in the day-to-day ICS policymaking process, and adoption would entail significant resource commitments to gain deep substantive knowledge of an unproven valuation construct and approach to risk quantification. The ICS would not provide meaningful incremental prudential benefits, beyond the utility already conferred by existing jurisdictional regimes. At the same time, deploying an unproven and unprecedented standard like ICS might subvert other important supervisory objectives, such as narrowing existing protection gaps and providing affordable insurance coverage for underinsured communities.

In terms of the potential impact of ICS on insurance market competitiveness, the CRO Council does not believe that the uniform application of a single global standard like ICS would promote or achieve a "level playing field" across insurers. The marketplace for several forms of insurance coverage is local and jurisdiction-specific, with both the insurance product marketplace and the surrounding regulatory regime evolving in an integrated and complementary manner over decades. There is an inherent technical challenge in seeking to develop uniform global risk capital treatments for products, activities, and investments that exhibit important local nuances. For example, within the US system, state guaranty associations play a vital risk-mitigating role in back-stopping product liabilities. These mechanisms are uniquely jurisdictional, with insurers liable for assessments based on their level of participation in a particular state marketplace.

Moreover, local jurisdictional rules serve as the primary anchor and focal point for solvency and capital management. Determining the assets that defease corresponding liabilities is primarily managed at an entity level, rather than from a global consolidated posture. Along the same lines, applying a dual regulatory system (i.e., if the ICS were implemented as a group capital regime alongside the current US state-based requirements for individual insurer entities) would sow confusion, create unwarranted complexity, and pose conflicts with existing US jurisdictional rules, including the regulation of insurance policy rates. ICS implementation might have legal implications affecting individual state regulation of an insurer's rates, given that the ICS is anchored in a total balance sheet approach based on the interdependence of consolidated group-wide assets, liabilities, regulatory capital requirements, and capital resources.

Existing US state-based regulations apply a "windows and walls" approach, which conservatively does not treat capital owned by individual insurers in a group as being interdependent, readily available, or fungible. Rather, state-based regulation ensures that each individual insurer, subject to multi-faceted requirements and oversight, maintains ample unencumbered capital to meet its obligations. In turn, state-based rate regulation emphasizes the assessment of capital adequacy of the individual legal entity issuing the policy, rather than the group as a whole. Applying the ICS as an additional capital requirement, as part of a dual regulatory system, is likely to create uncertainty and conflict with state-based rate regulation. Further, given that this dual regulatory system is envisioned to only apply to a subset of insurers within a market (i.e., Internationally-Active Insurance Groups, or IAIGs), the ICS would likely create an un-level playing field, given the local and jurisdiction-specific context within which insurers compete.

We believe the ultimate objective of the ICS process should be to build trust across group-wide supervisors in their respective approaches to enterprise-wide oversight, rather than to apply the same uniform and mechanistic rules to groups that operate globally. While a purpose of the ICS is to provide insurance supervisors globally with a "common language" for risk capital across cross-border groups, the demarcation of IAIGs as being subject to a distinct capital treatment could result in competitive hurdles relative to non-IAIGs operating in the same domestic market.

2. Please provide information on whether the ICS could create regulatory capital arbitrage opportunities or have procyclical effects, leading to increased volatility in U.S. insurance markets.

A stated objective of the ICS is to reduce regulatory capital arbitrage across jurisdictions, by applying the same requirements to all exposures and activities globally. However, this goal seems misplaced, given the local nature of many segments of the insurance product marketplace. A globally-uniform ICS might deliver a false sense of comparability, by disregarding local attributes and practices that impact risk exposure.

In certain respects, the ICS might exacerbate regulatory capital arbitrage, by incentivizing certain activities that do not align with an insurance group's internal view of economic risk. For example, MAV incentivizes groups to over-weight in high-grade Corporates (which are eligible assets under the MAV criteria) relative to, for example, long-term equity investments. Over-weighting in high-grade Corporates (or Sovereigns) would improve a group's ICS ratio, even though such a strategy might undermine a prudent ALM and diversification strategy that has been optimally designed to satisfy policyholder obligations.

The ICS might also inadvertently result in pro-cyclicality. The MAV is highly sensitive to changes in market conditions. If such market volatility were temporary and reversible, the ICS might indicate a synthetic and potentially distortive shortfall in reported capital levels. Namely, if compelling unwarranted assets sales, the ICS could undermine financial stability by exacerbating "fire sale" risks during a stress period. Such actions would undermine the inherent stabilizing role of insurance companies during turbulent markets, by subverting ALM and cash flow testing disciplines. As a general matter, the strict ICS adherence to "cash flow matching" in some cases would restrict the ability of insurers to optimize ALM over the long-term, to the detriment of policyholders. ICS capital requirements for market risk are also pro-cyclical, since the underlying calibration is independent of the current market environment and assumes a "stress on stress" calibration during market downturns periods.

The ICS might also amplify pro-cyclicality by driving "herding" or risk concentrations that align with the ICS incentive structure. For example, the NDSR charge, by penalizing holdings of Corporates, would artificially incentivize excess exposure to Sovereigns.

3. How should the FIO Study consider the potential effects of implementing the AM in U.S. insurance markets as compared to implementing the ICS? In addition, should the FIO Study consider the potential impact upon U.S. insurance markets if credit rating agencies were to accept the ICS as a global standard?

To help drive a successful outcome for the AM comparability assessment, the FIO study should compare the attributes of AM relative to the ICS across market cycles. The study should also compare how well AM and ICS, respectively, align with insurers' current practices in managing capital, which is primarily geared to optimizing capital adequacy on a jurisdictional basis, which for the US is based on the individual insurer meeting its obligations to policyholders and its solvency requirements. FIO should also consider the information value for supervisors, including consideration of the potential for the ICS to yield "false positives" and "false negatives" with respect to capital adequacy.

These comparisons should not solely involve a reductive quantitative side-by-side evaluation of the respective AM versus ICS ratios, but rather should encompass a broader, pragmatic analysis of the utility of each measure in addressing a given "real world" situation. As CROs, we consider not only the theory and concepts underlying a given risk measure, but also undertake a holistic assessment of a given measure's performance across a range of scenarios, as well as its informational value in the context of other tools, both quantitative and qualitative.

We recognize the appeal of a globally-comparable solvency metric among market participants, including rating agencies. However, we caution against a reliance on ICS to compare insurers' relative financial strength, given that the ICS is prone to presenting an inconsistent picture of financial strength across insurance groups, including both "false positives" and "false negatives" (e.g., overstating the deterioration in capital during temporary bouts of market volatility; and overstating the capital strength of insurers that invest heavily in Sovereigns, without due consideration of ALM and liability defeasance). Moreover, the scope of application of the ICS to IAIGs only does not align with how investors categorize the investable insurance universe. For example, a formally-identified IAIG would be subject to ICS, but could fall into the same peer group as a (potentially much larger) non-IAIG that only operates domestically.

4. What information should be considered in evaluating the impact of ICS implementation on the various business lines and the cost and availability of different product types in the U.S. insurance market?

The FIO study should apply scenario-based analysis, including the assessment of ICS under both stress and benign market conditions. As risk officers, we view it as essential to understand the implications of new systems under a spectrum of plausible conditions, and not be confined by the "availability bias" of only looking at a few recent years of historical observations. This analysis should apply an integrated view, taking into account not only the impact of ICS capital charges, but also of the valuation of insurance liabilities and MOCE.

The consideration of costs should encompass not only the potential for sub-optimal allocation decisions incentivized by the introduction of ICS capital charges and the MAV, but also the significant implementation outlays for adopting a completely new and unprecedented regulatory capital regime. Both the fixed and variable costs of ICS implementation would result in cost increases borne partly by consumers. As a reference point, the FIO study could look to the costs that European insurance groups faced in implementing Solvency II. Similarly, FIO might want to consider conducting a simple survey across US insurance groups to gauge roughly the prospective costs of implementing ICS.

The FIO study should view the impacts from the standpoint of consumer protection, in particular the risk that improperly calibrated treatments could deter the provision of economically-viable insurance coverage. Additionally, FIO should consider potential impacts not only to IAIGs, but also to purely domestic US companies. The study should assess how ICS implementation might impact these groups differently, as well as the effects on the overall US insurance market.

5. If the ICS were implemented in foreign jurisdictions where U.S. insurers operate, what effects could the ICS have on the ability of U.S. insurers to compete with local insurers and other international insurers in these overseas markets? How should FIO evaluate issues related to global competitiveness of U.S. insurers and potential adoption of the ICS by foreign jurisdictions?

There is an important distinction between supervisors applying ICS as a group regime, versus adopting ICS as a jurisdictional standard for locally-operating entities. It is of course within their authority for non-US regulators to adopt ICS as the basis for local solvency standards. While the flawed design of the ICS could result in negative impacts for the market and its consumers, there would not necessarily be adverse impacts to competition provided the regime is applied fairly across the insurance market.

The potential application of ICS by non-US supervisors as an additional group standard (i.e., in addition to the Group Capital Calculation or AM) underscores why the effort to demonstrate AM comparability is such a significant initiative. Should a US group be subject to the AM (by their state group-wide supervisor) and, in parallel, adhere to a redundant "shadow" ICS expectation (applied by non-US supervisors that do not trust/accept the AM), the group would be faced with both unwarranted costs and significant operational and risk management challenges. Further, we are concerned about the competitive disadvantages that would result if US insurers were subject to multiple (formal and "shadow") standards, whether those be based on the entity or sub-group level or the entire group, while peers operating in the same markets were only subject to one standard.

We view it as essential that the IAIS comparability assessment not succumb to an "anchoring bias"; namely, the assessment should not treat the ICS as an ideal benchmark that the AM should seek to replicate, but, rather, the AM should be evaluated based on its prudential utility in addressing solvency for the entities in US groups. We also believe that the optimal system for coordinated global oversight is for jurisdictional supervisors to defer to the group-wide supervisor's chosen approach to IAIG assessment, provided it can serve as an effective tool for assessing risks across the group and promoting sound risk management across the group. Absent a comparability determination, internationally-active groups would incur significant burden, costs, and sub-optimal capital allocation (i.e., "trapped capital").

6. Please provide your views on the following issues, as relevant to the FIO Study.

a. Data for FIO Study: The ICS has been developed with data provided by volunteer insurance groups. To what extent should FIO use data provided to FIO by individual insurers to conduct the FIO Study? In addition to data from specific insurers, are there any other relevant data sources that should be used to evaluate the ICS? If so, what other sources of quantitative and qualitative data would be available, including any data that could be representative of U.S. insurance practices and product types.

Data should be driven by the themes and topics of focus in the study. In other words, FIO should focus on gathering data that is directly instrumental to its pre-determined analytical objectives. In the interests of efficiency, we encourage FIO to coordinate its efforts with existing initiatives, such as that of the Federal Reserve's IPAC, to avoid duplicating efforts. FIO should aim to identify issues which can be assessed through targeted or tailored analysis. Performing a total balance sheet exercise is unwieldy and might not necessarily be relevant for examining particular technical issues.

As a first step, FIO should identify the areas of ICS that are germane to US insurers, but that have not yet been studied in an analytically rigorous manner. We encourage the exploration of market and sectoral data to illustrate broader trends or properties within insurance markets, with a view to how the ICS might distort existing practices. For example, an illustration of the depth of US corporate bond and commercial real estate markets could help to show why these asset classes are an important element of insurer ALM. Likewise, highlighting the depth and liquidity of US senior debt capital markets could be valuable in demonstrating why senior debt is an important and reliable avenue for insurers to recapitalize their operating entities under stress. We also think there is value in FIO applying hypothetical or stylized examples, which could be more tractable and less resource-intensive than trying to use real-world data.

FIO should work to complement and build around the IPAC study, as warranted. Ultimately, the IPAC results could become part of the FIO study's broader fact pattern, particularly regarding the potential adverse impacts on long-duration products under ICS. Indeed, the FIO study could provide a useful and prominent platform to help promote the IPAC work to a broader audience of stakeholders.

b. Market Effects from MAV: The reference ICS is based on a market-adjusted valuation methodology. What information should be considered in assessing MAV versus other valuation approaches and their potential effects on the insurance market in the United States, including consumers and insurers?

In comparing MAV to US statutory reporting, FIO should demonstrate both the US system's relative stability across the cycle, as well as its embedded conservatism, including (i) annual cash flow testing to adjust liabilities as needed based on changes to the risk profile and (ii) counter-cyclical buffers such as the Asset Valuation Reserve and Interest Maintenance Reserve. US statutory reporting balances the merits of relative stability over the cycle with appropriate risk-sensitivity, since it includes effective mechanics to adjust for deviations from initial assumptions. The analysis should also take into account the relatively greater significance of reserves as a portion of total financial strength under the US system, given the conservatism of the NAIC's statutory reserving methodology.

Since US insurers currently manage to the domestic statutory regime, a shift to a fundamentally different valuation basis would create differing incentives and disruptions to current practices and strategies. Insurers generally deploy a buy-and-hold strategy and incur limited liquidity risk, if applying disciplined ALM. As such, FIO should assess how short-term market volatility might increase ICS capital and reserve requirements for long duration products and understate capital adequacy during periods of temporary market distress, which could exacerbate financial strains. Additionally, it is important to emphasize that, even if ICS were not a binding constraint on capital held by the various entities in a group, its implementation would still have allocative effects by penalizing certain activities and practices, while rewarding others.

In particular, how should the FIO Study consider how MAV affects the following areas?

i. Changes to U.S. insurer investment behavior and ability to match asset-liability cash flows;

MAV introduces synthetic breakages between assets and liabilities. While the ICS incentivizes strict cash flow matching, sound insurer ALM and risk management does not necessarily entail nor necessitate such strict matching (as reflected, for example, in ICP 15, which recognizes that, given uncertainty in liability cash flows and the potential lack of assets with appropriate cash flow attributes, insurers are "usually not able to adopt a completely matched position"). Insurers test cash flows to ensure asset adequacy; however, policyholder interests are often best served by insurers applying a degree of judgment that might deviate from the exact, mechanistic matching of cash flows.

ii. Implications for product offerings and shifts in product mix for both life insurers and property & casualty insurers; and

As noted previously, we caution that ICS adoption would entail significant implementation costs that would be borne partly by customers. P&C insurers subject to ICS would need to undertake significant systems investments, without any meaningful impact on their overall solvency picture, since their generally shorter-duration assets and liabilities would not be meaningfully impacted by the application of MAV. Along these lines, FIO might want to perform a simple survey of US insurers to estimate the pro forma ICS implementation costs.

Additionally, the ICS applies a series of regulatory judgments and parameters (as, for example, hardwired within the technical criteria for determining MAV), which could drive herding behaviors, cliff effects, or artificial capital constraints. The ICS might drive a shift away from long-duration life insurance products (e.g., US-style life and annuity products with long-term guarantees and savings elements) that have met consumer needs, while being supported by existing effective risk management methodologies aligned with the current, conservative US statutory valuation and capital regime.

iii. Potential effects on insurers' role as a significant source of long-term investment and liquidity in the economy.

As noted above, US insurers generally deploy a buy-and-hold investment strategy and incur limited liquidity risk as they employ their disciplined ALM strategies. As a result, US insurers' business models serve as a significant source of long-term investment and liquidity in the economy. By introducing the greater volatility and pro-cyclicality described above, the ICS could undermine that role by interfering with US insurers' investment strategies and/or exacerbating potential "fire sale" incentives during turbulent markets.

Effective ALM strategies may also include long-term equities, including private, listed and real estate that over a sufficiently long period of time have performed as well as fixed income investments. The ICS Study should consider how these appropriate strategies are reflected in the asset and liability frameworks of the ICS relative to the AM.

c. Capital Requirement: The ICS capital requirement is based on a standardized framework, whereby the calculation of ICS required capital, including the risks and stresses, is defined. How should the FIO Study consider the following?

i. The extent to which jurisdiction-specific risks should be taken into account; and

The study should elucidate the numerous sources of difference across jurisdictions and markets, which is one of the core challenges in the effort to develop a uniform global standard like ICS. A fundamental problem with ICS is that its hardwired assumptions and parameters have become overly prescriptive and insufficiently tailored across markets. Jurisdictions can vary in terms of: asset class risks, properties, and performance (e.g., commercial real estate, infrastructure; private equity, private placement bonds); investment patterns and market depth (e.g., US corporate bond markets are notably deeper than in other jurisdictions); risk assessment processes and methodologies (e.g., NAIC designations); and insurance risk experience (e.g., the ICS 30% mass lapse assumption for retail life insurance products is an overbroad assumption).

A particular point of jurisdictional difference that the FIO study could help to elucidate is how the ICS spread risk charges (NDSR) asymmetrically affect US insurance groups relative to global peers. For example, through a hypothetical example, FIO could show that spread volatility (if short-term and mean reverting) does not affect long-term cash flows. The study could reference the historical ICS Field Testing data showing that US groups account for most of the NDSR reported across the global cohort. These onerous charges would undermine the role that US life insurers, in particular, play as investors in corporate debt markets.

Another area for potential research is the current ICS mass lapse assumption for retail insurance products, which is unrealistic for the US experience and should be differentiated based on historical lapse patterns. The negative impacts are apparent in high rate environments, unlike the time period during which the ICS has been developed. The FIO study could also play a constructive role both testing the impact of ICS capital charges across multiple economic environments and coordinating a collection of historical data to support development of a more reasonable assumption.

ii. The use of internal ratings for assessing credit risk exposures.

The use of both internal ratings and supervisory assessments are important to providing a more refined and tailored view of risk, by both expanding the scope of coverage beyond the NRSRO-rated universe (e.g., private placement bonds) and compensating for the misapplication of NRSRO ratings (e.g., for legacy securitization holdings).

To further illustrate the appropriateness of both internal ratings and supervisory credit assessments, the FIO study could evaluate a sample of representative transactions to show how internal ratings and NAIC designations align more closely with underlying economic risk exposure than NRSRO ratings. For legacy securitizations, the study could illustrate how the use of NRSRO ratings (which are designed to reflect only default probability, or first dollar of loss) significantly overstates the capital needed to absorb unexpected loss on an economic basis. NRSRO ratings, unlike NAIC designations, reflect neither the expected recovery (which can be very high, particularly on thicker tranches that have been downgraded by NRSROs), nor the discounted purchase price (i.e., the inherent loss absorption of buying the security below its fundamental value). For private placements, the study could illustrate how the application of an "unrated" ICS capital charge significantly overstates credit risk. The study could use either hypothetical examples or broader credit performance data on private placements to illustrate that their fundamental credit risk exposure is well below the levels implied by the "unrated" ICS charge.

d. Available Capital: The reference ICS measures available capital according to IAIS-established criteria and composition limits. The IAIS is also considering transitional arrangements during the monitoring period in order to ensure a smooth transition of the ICS as a PCR. How should the FIO Study consider the following?

i. Application of transitional arrangements during the monitoring period; and

ii. Implications for the fungibility of capital [13] under the ICS.

The ICS assumes that group-wide capital is completely fungible, which is a less conservative assumption than the AM. The AM, by respecting and building from legal entity regulatory capital requirements, is better aligned with real-world constraints on fungibility than the ICS is. Such realities should be of high importance when considering the appropriateness and ability of a group capital framework to serve as an effective tool for supervisory communication, coordination and cooperation.

e. Jurisdictional Flexibility: The reference ICS recognizes a limited number of areas for national discretion, such as senior debt as qualifying capital. Should the FIO Study evaluate any further application of jurisdictional flexibility for ICS implementation?

In general, it is inherently challenging to develop uniform global methodologies covering every form of risk exposure and activity. Jurisdictional flexibility is therefore important, as strict limits and hardwired parameters could inhibit new product design, market development, and investment classes. As an example, it would have been difficult a few decades ago to envision the currently sizable cyber risk market.

As an area for further research, the FIO study could illustrate how senior debt is an important pathway to raising capital for operating entities during periods of stress. For example, the study could cite historical examples and data supporting how insurance groups have been able to access debt capital markets during periods of market volatility.

7. Please provide any views regarding the following additional issues, as they relate to the FIO Study.

a. What data and input from market participants should be taken into consideration?

As noted previously, we believe that, for efficiency purposes, the FIO study should seek to leverage existing data, research, and analysis, including historical ICS Field Test results, the forthcoming IPAC study, and market-wide data. We believe it is important to limit data requests or calls specific to the study, given competing supervisory data requests to support the prudential standard-setting process. For example, the IAIS is already requesting data for both the MP and global systemic risk monitoring, which has ramped up during the COVID pandemic. FIO will face challenges in seeking to collect sufficient data across a broad cohort of companies, if such requests were not already tied to existing data collection exercises.

b. Describe any data or data services that independent third parties could provide for purposes of the FIO Study.

The choice of datasets should depend on the analysis and themes covered in the report; namely, we encourage FIO to first define the intended contents of the study and then determine the appropriate data required to fulfill those research objectives.

c. For the purposes of the FIO Study, would a "point in time" analysis be appropriate or would another time frame be more relevant for determining the implications?

We encourage FIO to perform its analysis across scenarios, including both benign periods during which risk exposures might build, as well as stressed environments during which the ICS may be prone to pro-cyclicality. However, we do not encourage bespoke stress testing for the purposes of the FIO study, given the resource-intensity of such an exercise. Additionally, insurer liabilities in some cases run for decades, so we believe there is value in considering impacts across a similar timeframe.

8. How should the FIO Study inform FIO's engagement on the IAIS economic impact assessment of the ICS?

FIO should seek to ensure that the study is a meaningful exercise of sufficient breadth and scope to assess the potential for unintended consequences to the industry, consumers, and other stakeholders. Further, it is critical that both the design of the study and analysis of the results be objective, rather than treated as a pro forma step toward finalization of the framework. To help accomplish these objectives, we believe FIO should collaborate with the NAIC and Federal Reserve to approach engagement on the study from a united US perspective, to strengthen the ability to secure the following design elements:

* The study assesses the performance of the ICS under a range of market conditions (benign and stress).

* The study should explore the externalities the ICS could give rise to, if implemented as a prescribed capital requirement for IAIGs (in all markets they exist - including the US), including implications for risk management, product pricing and insurance coverage, competition, the provision of funding to capital markets (with emphasis on longer tenor assets and asset allocation by insurers), and the potential for conflict with existing regulatory tools and requirements.

* The study should address the potential for the ICS to create tension or conflict between jurisdictional regulatory requirements and tools (entity and group level) for insurers and supervisors.

* The study should consider the aspects noted above for alternative approaches to an ICS as well - particularly the AM.

9. How has the COVD-19 pandemic informed your views on the issues discussed in this Notice?

COVID has truncated the Monitoring Period timeline, as 2020 has not been as productive in working towards achieving the milestones laid out in the Abu Dhabi agreements and engaging with stakeholders - both industry and perhaps more importantly, the frontline supervisors for whom ICS is intended to serve as a resource. Both policymakers and insurers necessarily re-prioritized efforts to respond to COVID (e.g., through increased monitoring and data calls). We believe strong consideration should be given to elongating the Monitoring Period by one year, given these COVID-related disruptions and the importance of ensuring the final product is fit for purpose and will not give rise to unintended consequences.

Additionally, the volatility experienced at the end of Q1 2020 provides an opportunity to study the performance (and potential pro-cyclicality) of ICS during a period of severe but temporary market turbulence. While the IAIS collected a view of the ICS balance sheet as of this valuation date, we believe that limited data point alone is insufficient for drawing any substantive conclusions. Consideration of how required capital charges were impacted is a critical component that was not captured. Similarly, the IAIS did not assess how the measures of available and required capital evolved - and the implications for management and supervision of insurers - over the course of the year as markets reacted to the pandemic. It is also worth considering the counterfactual of how central bank interventions during the height of the pandemic significantly tempered market turbulence and, in turn, even more severe volatility in market-based measures such as ICS.

10. Please provide any other comments on the issues discussed in this Notice.

We look forward to continued engagement with FIO, beyond the points raised in this letter, as the design, substance, and process for the study takes shape. The CRO Council is prepared to provide ongoing guidance and perspective as FIO further shapes the content of the analysis.

Sincerely,

Jonathan Porter

Chair of North American CRO Council

* * *

The notice can be viewed at: https://beta.regulations.gov/document/TREAS-DO-2020-0019-0001

TARGETED NEWS SERVICE (founded 2004) features non-partisan 'edited journalism' news briefs and information for news organizations, public policy groups and individuals; as well as 'gathered' public policy information, including news releases, reports, speeches. For more information contact MYRON STRUCK, editor, [email protected], Springfield, Virginia; 703/304-1897; https://targetednews.com