Insurers: Rating Methodology

Transcription

1 Criteria Insurance Request for Comment: Insurers: Rating Methodology Financial Services: Emmanuel Dubois-Pelerin, Global Criteria Officer, Paris (33) ; Corporate & Government Ratings: Colleen Woodell, Chief Credit Officer, New York (1) ; Primary Credit Analysts: Rodney A Clark, FSA, New York (1) ; [email protected] Rob Jones, London (44) ; [email protected] Mark Button, London (44) ; [email protected] Secondary Contacts: Matthew Carroll, CFA, New York (1) ; [email protected] John Iten, New York (1) ; [email protected] Lotfi Elbarhdadi, Paris (33) ; [email protected] Connie Wong, Singapore (65) ; [email protected] Angelica Bala, Mexico City (52) ; [email protected] Gregory Gaskel, New York (1) ; [email protected] Karin Clemens, Frankfurt (49) ; [email protected] Damien Magarelli, New York (1) ; [email protected] Kevin Ahern, New York (1) ; [email protected] Michelle Brennan, London (44) ; [email protected] David Laxton, London (44) ; [email protected] Table Of Contents I. INTRODUCTION II. SCOPE OF THE PROPOSED CRITERIA III. PROPOSAL SUMMARY IV. SPECIFIC QUESTIONS FOR WHICH WE ARE SEEKING A RESPONSE JULY 9,

2 Table Of Contents (cont.) V. RESPONSE DEADLINE VI. IMPACT ON OUTSTANDING RATINGS VII. METHODOLOGY A. Calibrating The Ratings B. Determining The Ratings C. Assessing The Business Risk Profile C1. Deriving The Business Risk Profile C2. Insurance Industry And Country Risk Assessment (IICRA) C3. Competitive Position D. Assessing The Financial Risk Profile D1. Deriving The Financial Risk Profile D2. Capital And Earnings D3. Risk Position D4. Financial Flexibility E. Modifiers And Caps To The Indicative SACP Or GCP E1. ERM And Management Score E2. Liquidity E3. Fixed-Charge Cover Test E4. Rating An Insurer Above The Sovereign Rating Or T&C Assessment F. Support Framework F1. Rating Insurance Subsidiaries Of Insurance Groups F2. Assigning ICRs To Nonoperating Holding Companies F3. Assigning Issue Ratings VIII. GLOSSARY IX. RELATED CRITERIA AND RESEARCH JULY 9,

3 Criteria Insurance Request for Comment: Insurers: Rating Methodology I. INTRODUCTION 1. Standard & Poor's Ratings Services is requesting comments on changes it is proposing to its criteria for rating insurers. Our intention is to enhance the transparency of our methodology on insurers globally. The proposed criteria introduce a ratings framework comprising a business risk profile and a financial risk profile. They also put forward new rating factors and subfactors to assess the impact of industry and country risks, prospective capital adequacy, and risk position. The aim is to clearly and in considerable detail specify the factors and subfactors of the analysis, and to show how they combine into rating outcomes. 2. The criteria propose quantitative and qualitative metrics for evaluating subfactors for these rating factors: industry and country risk, competitive position, capital and earnings, risk position, liquidity, and financial flexibility. No changes are proposed to the enterprise risk management factor or to Standard & Poor's risk-based capital adequacy model. II. SCOPE OF THE PROPOSED CRITERIA 3. The proposed criteria apply to all global-scale foreign currency and local currency long-term issuer credit, financial strength, and financial enhancement ratings on insurers in the business of life, health, and property/casualty (P/C; known as non-life outside of the U.S.) insurance and reinsurance sectors. For most companies, the three types of ratings are identical under the current and proposed criteria. The criteria exclude ratings on bond insurers, insurance brokers, insurers that are starting up or are in run-off, and mortgage and title insurers. Public information ("pi") ratings are out of the scope of this Request for Comment (RFC). III. PROPOSAL SUMMARY 4. The proposed criteria for insurance ratings constitute specific methodologies and assumptions under Standard & Poor's "Principles Of Credit Ratings," published on Feb. 16, The methodology we propose consists of two key steps (see chart 1): assessing the stand-alone credit profile (SACP) and then extraordinary government or group support. Once a rated insurer's group member status and the likelihood of extraordinary support are evaluated, then the criteria assign the insurer's issuer credit rating (ICR) as a function of the group credit profile (GCP) and, for government-related entities (GREs), the rating on the government. In some cases, set out in 21 to 24, the proposed criteria allow for the assignment of an SACP or GCP that is one notch higher or lower than the criteria for the SACP or GCP imply. 6. The assessments of the SACP and GCP rest on the same eight rating factors: Insurance industry and country risk assessment (IICRA), JULY 9,

4 Competitive position, Capital and earnings, Risk position, Financial flexibility, Management and corporate strategy, Enterprise risk management (ERM), and Liquidity. 7. An SACP, a GCP, and the ratings on an insurer are subject to our country risk assessments. In addition, foreign currency ratings on domestic unsupported insurers are no higher than our transfer and convertibility (T&C) assessments. JULY 9,

5 JULY 9,

6 IV. SPECIFIC QUESTIONS FOR WHICH WE ARE SEEKING A RESPONSE 8. Standard & Poor's is seeking market feedback on its proposed criteria and responses to the following questions: Do the criteria incorporate the key factors affecting an insurer's creditworthiness? Do you agree with the main variables for assessing the different factors? If not, what is missing and what is redundant? Are we sufficiently clear and transparent about how we explain the proposed process for assigning ratings, and standards for evaluating and weighting the proposed rating factors? If not, what areas would benefit from greater clarity? Do you agree with the way the proposed insurance industry and country risk assessment (IICRA) score is reached and would affect insurers' ratings? If not, what alternatives would you propose? Do you agree with the proposed way that liquidity, enterprise risk management, and management are scored and how they would affect ratings? V. RESPONSE DEADLINE 9. We encourage all market participants to submit written comments on the proposed criteria by Sept. 9, Please send them to [email protected]. After the deadline, we will review the comments and publish the criteria. VI. IMPACT ON OUTSTANDING RATINGS 10. We expect any change to our global distribution of insurer ratings to be modest. The review may lead to adjustments to some insurance company ratings. We expect the significant majority of ratings to remain unchanged or move by no more than one notch. VII. METHODOLOGY A. Calibrating The Ratings 11. We calibrate our insurance ratings criteria based on our analysis of the history of defaults, the impact of various financial and economic crises on insurance company creditworthiness, the credit strength of the insurance sector compared with that in other sectors, and on Standard & Poor's framework for the behavior of our credit ratings over time through economic cycles. We outline our framework in three articles: "Understanding Standard & Poor's Rating Definitions," published on June 3, 2009; "Credit Stability Criteria," May 3, 2010; and "The Time Dimension Of Standard & Poor's Credit Ratings," Sept. 22, Insurance companies are typically highly regulated, and in general the regulatory framework has been effective. To protect policyholders, insurance companies are normally required to hold levels of capital in excess of required "solvency margins" to offset potential losses. JULY 9,

7 13. Consequently, although Standard & Poor's insurance ratings span the entire rating scale, there are a greater proportion of ratings at the higher end than in most other sectors. Furthermore, the median rating of the universe of rated insurers is higher than in all sectors except governments. Of insurer financial strength ratings, 93% are currently investment grade (including public and confidential ratings). Of those, 52% are in the 'A' category, while 24% are in the 'AA' category, compared with about 15% and 2%, respectively, for nonfinancial corporate issuers. Our rated universe is heavily concentrated in developed countries and midsize to large insurers, and it is likely that many of the unrated insurers would fall in lower rating categories. 14. The main sources that we have used to review the history of insurance company defaults are Standard & Poor's default studies (see "2011 Annual Global Corporate Default Study And Rating Transitions," March 21, 2012), which covers the performance of Standard & Poor's insurance ratings, both in terms of transition and default, over the period 1981 to We note that creditworthiness in this heavily regulated sector appears to be sustainable during periods of economic stress. Default rates have increased during periods of stress, such as economic downturns, or following major catastrophes, but have remained relatively low. We note that the rated average default rate over the study period is 0.41%, which is the lowest of any sector in the study. 15. Our criteria are informed by several periods of heightened stress that resulted in an increased number of significant insurance company failures. The periods of stress were more industry-specific than macroeconomic: : A number of predominantly casualty insurers, including Mission Insurance Co. and Transit Casualty Insurance Co., became insolvent as loss reserves proved deficient following a period of inadequate pricing industrywide (source: "Failed promises: Insurance company insolvencies: A report by the Subcommittee on Oversight and Investigation of the Committee on Energy and Commerce," U.S. Congress, 1990; : Several significant life insurers, including Executive Life Insurance Co., Mutual Benefit Life Insurance Co., and Confederation Life Insurance Co., failed due to a combination of illiquid asset concentrations and run-on-the-bank scenarios; 2000: Japanese insurers, including Chiyoda Mutual Life Insurance Co. and Kyoei Life Insurance Co., voluntarily entered rehabilitation, as guaranteed rates of interest on savings products were no longer sustainable given low interest rates in Japan (source: "Why Some Japanese Insurers Are Failing," Towers Perrin, 2001); and : Several P/C insurers and reinsurers failed, including Mutual Risk Management Ltd., Trenwick Group Ltd., Globale Rückversicherungs AG, and Converium Reinsurance (North America) Inc., predominantly due to deficient reserves for casualty lines following a period of inadequate pricing industrywide, and weak risk management. 16. The proposed criteria globally address the issues that caused these failures, including, among other areas (1) new liquidity metrics, (2) capital metrics that focus more on asset-liability risks, (3) IICRA metrics that take into account industrywide pricing adequacy, and (4) a larger role for ERM for companies with complex risks. 17. The global financial crisis did not trigger a wave of insurance life and P/C defaults. In fact, no significant insurer rated by Standard & Poor's defaulted due to the financial crisis, other than in the bond insurance and mortgage insurance sectors, both of which are outside the scope of the proposed criteria (see "Bond Insurance Rating Methodology And Assumptions," Aug. 25, 2011, and "U.S. Mortgage Insurer Sector Outlook Remains Negative--And The Clock's Ticking," March 1, 2012). However, American International Group Inc. would have failed without a rescue by the Federal Reserve Bank of New York. Although the company's problems fell largely outside of its insurance businesses, JULY 9,

8 we believe this criteria proposal, together with our revised bank and collateralized debt obligation criteria (see the Related Criteria And Research section for bank criteria, and "Global CDOs Of Pooled Structured Finance Assets: Methodology And Assumptions," Feb. 21, 2012), would have captured them. B. Determining The Ratings 18. SACP and GCP are as defined in "Stand-Alone Credit Profiles: One Component Of A Rating," published Oct. 1, 2010, and "Group Rating Methodology And Assumptions," published Nov. 9, The proposed criteria determine an operating company insurer's rating in six steps (for the scoring, see table 2): The business risk profile is derived from the combination of the scores for the relevant IICRA and the insurer's competitive position. The financial risk profile is derived from the combination of the insurer's scores for capital and earnings, risk position, and financial flexibility. The anchor is derived from the combination of scores for the business and financial risk profiles according to table 1 unless near-term and present default risk leads to a rating conclusion of 'D', 'SD', or 'CC' based on our ratings definitions. The indicative SACP or GCP is equivalent to the anchor, unless it is modified by the ERM and management score and by peer comparisons according to 21 to 24. The SACP is equivalent to the indicative SACP and the GCP is equivalent to the GCP, unless the liquidity, fixed-charge coverage, and sovereign risk tests imply a lower SACP. The ICR results from the combination of the SACP and of the support framework, which determines the extent of uplift, if any, for government or group support. Table 1 Anchor Financial risk profile (from table 12) Business risk profile (from table 3) aa+ aa aa- a+ a a bbb+ bbb- bb b+ 2 aa- aa- aa- a+ a a bbb bb+ bb b+ 3 a+ a+ a a- a- a- bbb bb+ bb- b+ 4 a a- a- bbb+ bbb+ bbb+ bbb- bb bb- b 5 bbb+ bbb+ bbb bbb bbb bbb- bb+ bb b+ b 6 bbb- bbb- bbb- bb+ bb+ bb+ bb bb- b b 7 bb- bb- bb- bb- bb- bb- b+ b b b- or lower Note: An issuer credit rating of 'AAA' is possible for example if the ERM and management score is '1' (see tables 19 and 20) or if the peer-based comparison leads to a one-notch adjustment (see 21). 19. A nonoperating holding company rating, under the proposed criteria, is assigned by notching down from the group's GCP, typically by a maximum of three notches. 20. GCPs and SACPs are assessed based on the same eight proposed rating factors in 6, and the assignment of the GCP follows the steps in 18. The scope of the GCP analysis is the entire group. By contrast, for a group member, the scope JULY 9,

9 of the SACP would be the entity itself or, if it owns subsidiaries, the subgroup. Table 2 Factor scoring Factor Strongest score Weakest score Business risk profile 1 7 IICRA 1 6 Competitive position 1 6 Financial risk profile 1 10 Capital and earnings 1 8 Risk position 1 4 Financial flexibility 1 4 ERM and management 1 5 Liquidity 1 5 IICRA--Insurance industry and country risk assessment. Throughout these proposed criteria, a score is "worsened" or "weakened" when the number increases (e.g. from '2' to '3'), and "improved" or "strengthened" when the number decreases (e.g. from '4' to '3'). 21. Under the proposed criteria, the assessment of an insurer's SACP and GCP is refined by up to one notch in either direction to reflect elements not captured elsewhere in the proposed framework. Such an adjustment considers the insurer's relative credit standing among peers, either through a holistic analysis of the eight rating factors (except liquidity) or by identifying below- or above-average vulnerability to event-related risks (often referred to as tail risks) or sustained, predictable operating and financial outperformance or underperformance. 22. Peers are insurers in the same sector as the insurer under consideration for global reinsurers, global marine protection and indemnity (P&I) insurers, global trade credit insurers, and global multiline insurers. 23. For other insurers, peers are rated insurers operating in the same country and P/C or life sector in the sense of table 10 and box 2. If the peer group contains a small number of rated peers, it may be broadened to include rated or unrated insurers in markets with the same IICRA score (or, if still insufficient, similar IICRA scores). 24. If the peer group is overly large, it may be delineated according to business line, such as in the U.S. where the P/C JULY 9,

10 insurance sector can be delineated between personal and commercial lines. C. Assessing The Business Risk Profile C1. Deriving The Business Risk Profile 25. The business risk profile (BRP) measures the risk inherent in the insurer's operations and therefore the potential sustainable return to be derived from those operations on a scale from '1' (excellent) to '7' (weak). 26. The BRP is based on the IICRA specific to the insurer and on the insurer's competitive position according to table Relatively low-risk product offerings or target markets with favorable competitive dynamics can strengthen the BRP score by one category. We would apply this adjustment infrequently, and only in cases where a majority of the insurer's liabilities exhibit a lower BRP than that of peers operating in the same sectors (e.g., health, life, or P/C) and countries. Some examples of such profiles include the following: Focus on products with meaningful risk-sharing features, such as participating whole life insurance and some with-profit products with minimal investment return guarantees, provided that the insurer has demonstrated the willingness and ability to share adverse experience with policyholders in spite of the commercial implications. Avoidance of markets where high-risk "secondary guarantees" have become prevalent, such as no-lapse guarantees on universal life insurance, or living benefit riders on variable annuity contracts; Focus on niche or underpenetrated markets with few competitors, effective barriers to entry, and sustainably strong margins. JULY 9,

11 C2. Insurance Industry And Country Risk Assessment (IICRA) 28. The proposed IICRA addresses the risks typically faced by insurers operating in specific industries and countries and is generally determined at a country or regional level. For example, we expect to assign an IICRA to the Canadian P/C sector, one to the Australian health sector, and one to the Japanese life sector. The IICRA anchors our analysis of an insurer's BRP, as industry and country risks are closely linked with the analysis of competitive position. A specific analysis is proposed for four global sectors (see 36 to 38). 29. Since the various subfactors all contribute significantly to an insurer's BRP, the IICRA score applicable to each industry or country combination is derived by applying Table 5 where columns indicate the average of the four country-related subfactor scores, and the rows the average of the five industry-related subfactor scores. Additionally, the IICRA score is no stronger than '4' if P/C, health, or life insurance premiums comprise less than 1.5% of GDP because such low penetration rates indicate that the insurance market is at a less mature stage of development. This restriction holds except for countries where GDP is highly influenced by exports, such as Hong Kong and Taiwan. 30. Four country-related subfactors--economic, political, financial system risk, and payment culture and rule of law--are scored on a scale from '1' to '6'. The other five, which are industry-related subfactors, are scored positive or '1', neutral or '3', or negative or '6'. For the four sectors discussed in 36to 38, each of the four country-related subfactors is assigned a score of '2'. This is an approximation of the global average of the country-related subfactor scores of the countries where these sectors' participants operate. 31. Table 4 shows how we would identify and score the proposed IICRA subfactors: JULY 9,

12 Economic risk, Political risk, Financial system risk, Payment culture and the rule of law, Return on total capital, Product risk, Barriers to entry, Insurance penetration trends, and Institutional framework. 32. The first four subfactors are country risks that affect all industries including insurance while the last five, though influenced by country risks, are specific to the insurance industry. The first three assessments are drawn from Standard & Poor's sovereign and bank industry criteria (see table 4). The inclusion of those and of the fourth subfactors reflects that: The industry's revenue and profitability dynamics are highly sensitive to the local economic environment. The industry is typically highly regulated. The industry is dependent on the banking sector for the transmission of money, on the provision of loans and facilities and, both with respect to fixed-income investment instruments and to its own financing, on deep and liquid debt capital markets. The industry is affected by the quality of the legal framework and of the judicial system. JULY 9,

13 33. Under the proposed criteria, an insurer operating in a single country and single insurance industry sector is assigned the IICRA score associated with that country and sector. 34. For insurers operating in more than one country or sector, we assign a weighted-average IICRA score by calculating the rounded-off average of the IICRA scores for the insurer's country and sectors, weighted by its gross premiums. We combine IICRA scores from the insurer's main markets to cover at least 90% of its business by premiums, up to 20 countries. 35. In rare cases, and when the averages of the country- or industry-related subfactors described in 30 or, for a given insurer, the premium-weighted average of its relevant IICRAs, falls within 0.2 of a cutoff point, the assessment also JULY 9,

14 factors in the directional trend of the overall IICRA. In such cases, the IICRA may be modified by one category. For example, of the three averages in this paragraph, one might fluctuate from one year to the next from 3.4 to 3.6. We will then assess whether, over the coming five years, the directional trend points more clearly to a '3' versus a '4'. Table 5 IICRA Assessment Country risk* Industry risk* *Equally weighted average of the related subfactor scores. 36. Insurers operating in the P/C reinsurance, life reinsurance, trade credit insurance, and marine protection and indemnity (P&I) sectors are assigned the sector's global score. This is because they typically write business in multiple countries around the world, resulting in high levels of geographic diversification. In addition, the domicile of the insurer has relatively little impact on the aggregate industry and country risks it faces. As indicated in 30, for each of these four sectors, a single, global IICRA applies and incorporates the '2' score for the four country-related subfactors. 37. However, if based on premiums an insurer or reinsurer in these four sectors focuses on a single country or region, an IICRA is applied at a country or regional level using the process that applies to all other insurers. 38. For the purpose of the proposed criteria, P/C reinsurance includes certain large commercial and industrial business lines that have similar characteristics to reinsurance because risks are commonly underwritten on a subscription or coinsurance basis. 1. Economic, political, and financial system risk 39. The economic, political, and financial system risk scores draw on our sovereign criteria for the first two elements and on our Banking Industry Country Risk Assessment (BICRA) criteria for the last element. The economic score in IICRA reflects the sovereign economic score as well as sovereign monetary and external scores and the BICRA score for external imbalances. The financial system risk score reflects both the BICRA's banking industry risk score, with additional weight given to the breadth or narrowness of domestic capital markets, and the domestic private sector's access to external funding. 2. Payment culture and rule of law 40. The payment culture and rule of law is another factor that influences an insurer's performance, given how important contractual arrangements are to this industry. The assessment of this subfactor addresses the predictability of the legal framework. The analysis is informed by external indicators, such as the World Bank's governance indicators for the rule of law, ease of enforcing contracts and control of corruption, and Transparency International's corruption perceptions index. JULY 9,

15 3. Return on total capital (ROTC) 41. The criteria propose to assess ROTC (see definition in glossary) on the aggregate profits and the aggregate total capital of at least 60% of industry participants by premiums (including all rated participants), or all rated participants, whichever is greater (see table 4). 42. Where we have insufficient public data to meet the 60% threshold, we typically assess the subfactor based on available evidence. 43. For markets where ROTC is either not available, volatile due to the influence of unrealized investment gains or losses, or low due to very high levels of capitalization, table 6 applies. 44. If there is insufficient evidence to form an opinion or the available evidence suggests excessive risk-taking is taking place, the score is "negative." 45. Note that the proposed criteria treat asset allocation and capitalization policy as entity-specific features. Table 6 Alternative Metrics For Assessing ROTC* Subfactor / score 1 (positive) 3 (neutral) 6 (negative) Life insurance New business margin (see glossary) The average margin over the past five years is 2% or higher The average margin over the past five years is between 2% and 1% or lower The average return over the past five years is 1% or lower Return on assets or prebonus pretax earnings/total assets* For either of the two metrics, the insurer's average return over the past five years is 1% or higher For either of the two metrics, the insurer's average return over the past five years is between 1% and 0.5% For either of the two metrics, the insurer's average return over the past five years is 0.5% or lower P/C insurance Return on revenue The average return over the past five years is 12% or higher The average return over the past five years is between 12% and 5% The average return over the past five years is 5% or lower *This alternate metric is used only if the new business margin metric is not available. Note: The metrics in this table are defined in "Assumptions For Quantitative Metrics Used In Rating Insurers Globally," published on April 14, P/C--Property/casualty. ROTC--Return on total capital. 46. In our opinion, risk taking is excessive if we perceive that any of the following conditions exists: Insurers have relaxed their underwriting standards. For example, premiums, prices, or policy terms and conditions have been or are being significantly reduced; or new and unproven products have been introduced and are growing rapidly; Mis-selling risk is heightened; for example, policy lapse rates are unusually high or policyholders are filing, or are expected to file, compensation claims for products sold to them; Commissions to intermediaries have significantly increased; or Premiums are insufficient to achieve long-term profitability. 47. Where the historic average could misrepresent the long-term performance we expect from the sector, we may base our assessment on our expectations. This is appropriate, in our view, when the average is skewed by, for example: The history of catastrophic events: for example, if the occurrence of catastrophic events has proved abnormally high (or low) over the past five years, we base the assessment on our normalized expectations; "Hard" or "soft" insurance markets (see glossary) that are unlikely to persist; for example, if markets have been predominantly hard over the past five years, we base the assessment on our expectations of market conditions JULY 9,

16 which may include lower premium rates or weaker terms and conditions; or The investment return contribution to ROTC, or to metrics in table 6, is unusually high or low because of interest rates or realized or unrealized investment gains. For example, we may adjust interest income received to reflect expected interest rates and disregard realized and unrealized investment gains to the extent that they exceed our expectation. 4. Product risk 48. Some product-specific elements can cause ROTC (or, if ROTC is not available, metrics in table 6) to be more volatile. For example: Property insurance underwriting results may be materially affected by catastrophes. Casualty insurance underwriting results may be materially affected by unpredictable settlements, for example, where legal systems include jury-awarded or punitive damages, or where claimant compensation arrangements or liability laws change frequently. P/C underwriting results may be materially affected by fraud. In another example, material marketwide life insurance asset-liability mismatch risk may exist, such as: variable annuities with living benefit guarantees; long-term care insurance; no lapse-guarantee universal life; insurance liabilities backed materially by equities (by management choice or because fixed-income instruments of sufficient duration are not sufficiently available); or where low or negative spreads exist due to current interest rates at or below contractual guaranteed rates. 49. Each of the sources of volatility described in the previous paragraph is assessed, for ROTC, or for the metrics in table 6, as high risk, moderate risk, or low risk; and we may identify and assess further industry- or country-specific sources of volatility stemming from product risk. In addition, if catastrophe risk is comprehensively reinsured, that source of volatility is not assessed as high risk. 50. The product risk score is "positive" if each of the sources of potential volatility in ROTC, or in the metrics in table 6, is assessed as low risk. 51. The score is "negative" if any of the sources of potential volatility is assessed as high risk. 52. In all other cases, product risk is scored "neutral." 5. Barriers to entry 53. Insurance is typically prudentially regulated, usually resulting in at least moderate barriers to entry. These barriers may be legal, regulatory, or operational (see 55 to 57). They are assessed as either high, low or, in the case of operational barriers, moderate. 54. If any of these barriers is assessed as high, the overall score is "positive," but the overall score is only "negative" if all barriers are assessed as low. In all other cases, the overall score is "neutral." 55. Legal barriers are assessed as high where access is limited to named insurers under the law or as part of government policy in the relevant industry-country combination. 56. Regulatory barriers are assessed as high where regulatory practices involve either exceptionally demanding or lengthy JULY 9,

17 procedures for licensing new insurers. 57. Operational barriers are assessed as high if availability is low and costs are high for resources such as management, staff, systems, data, and sources of distribution in the relevant industry-country combination. Operational barriers are assessed as low if availability is high and costs are low for resources such as management, staff, systems, data, and sources of distribution in the relevant industry-country combination. 6. Insurance penetration trend 58. Trends in insurance penetration enable us to assess the potential for the industry to grow, relative to overall economic growth, which is factored in the economic risk subfactor. The assessment is based on the growth or contraction of the ratio of life and P/C insurance premiums as a proportion of GDP. Where GDP growth is volatile, where possible we may remove volatile components (e.g. non-oil GDP growth may be a more suitable metric in certain countries). In rare cases, the assessment is adjusted to exclude offshore business. 59. The subfactor score is "neutral" for the industries assessed at a global level, i.e., P/C reinsurance, life reinsurance, trade credit insurance, and marine P&I. However, if a major industry development (e.g. a regulatory change) results in substantially increased or decreased product demand, the score would be "positive" or "negative," respectively. 60. The subfactor score is "positive" if we expect insurance penetration to grow significantly over the next three to five years. Typically, significant growth would correlate with more than 10% growth in the ratio of insurance premiums to GDP (for example from 2.0% to over 2.2%). However, if, in our view, growth is being achieved through excessive risk-taking, the score would be reduced to "neutral" (see 46). 61. The subfactor score is "negative" if we expect the ratio of insurance premiums to GDP to reduce significantly over the next three to five years. Typically, a significant reduction in the ratio would correlate with more than a 5% reduction in the ratio of insurance premiums to GDP (for example from 2.0% to under 1.9%). 62. In all other cases, insurance penetration is scored "neutral." In such cases, no penalty is incurred for excessive risk taking, which the ROTC subfactor already addresses. Examples of excessive risk taking are provided in Where we believe the historic average misrepresents our expectation of the sector's long-term trend, we base our assessment on the expected level. 7. Institutional framework 64. The strength of the institutional framework chiefly depends on regulatory oversight of the industry. Therefore, the score is "positive" if our assessment of regulatory oversight is strong, "neutral" if it is moderately strong, and "negative" if it is weak. The score is, however, reduced by one category (for example, from positive to neutral) where we observe a clear deficiency in the standards of either governance or transparency within the industry-country combination. 65. Regulatory oversight is assessed based on how sophisticated and effective the authorization and ongoing supervision requirements of insurers are in the relevant industry-country. Elements in the assessment include the depth and frequency of monitoring of insurers and the regulator's track record of intervening to reduce or mitigate the effects of insurer failures. 66. Only a few industry-country combinations are likely to have strong regulatory oversight under the proposed criteria. JULY 9,

18 Weak regulatory oversight assessments would be more frequent, often associated with emerging markets. If regulatory oversight is not assessed as strong or weak, it is assessed as moderately strong. 67. The proposed criteria assess governance standards by evaluating the balance of stakeholder interests among owners, managers, lenders, and policyholders. Corporate governance that is transparent, prudent, and independent of undue external influences lowers the risk of an insurance industry. Conversely, opaque, imprudent governance that does not materially constrain those external influences increases that risk. 68. The proposed criteria assess transparency by evaluating the frequency and timeliness of reporting, and the quality and standardization of financial reports. The criteria examine the quality of accounting and disclosure standards, including whether an insurance industry has adopted International Financial Reporting Standards (IFRS), U.S. generally accepted accounting principles (GAAP) or publicly available comprehensive regulatory returns. The assessment is also informed by the extent and effectiveness of a country's auditing requirements. C3. Competitive Position 69. The proposed criteria assess the level of an insurer's competitive position under six subfactors scored "positive," "neutral," or "negative" (see table 7): Operating performance, Differentiation of brand or reputation, Market share, The level of controlled distribution channels, Geographic diversification, and Other diversification. 70. Table 8 shows how the proposed ratings framework for insurers uses these subfactors to assess competitive position on a scale from '1' (extremely strong) to '6' (weak). 71. Underperformance or outperformance, as defined in table 9, influences four of the subfactors. JULY 9,

19 JULY 9,

20 JULY 9,

21 Table 8 Competitive Position Assessment Score and assessment* What it means Guidance (see table 7) 1 (extremely strong) An insurer's business operations make it significantly less vulnerable to adverse operating conditions than the IICRA score indicates. 2 (very strong) An insurer's business operations make it somewhat less vulnerable to adverse operating conditions than the IICRA score indicates. 3 (strong) An insurer's business operations are representative of the IICRA score. 4 (adequate) An insurer's business operations make it somewhat more vulnerable to adverse operating conditions than the IICRA score indicates. 5 (less than adequate) An insurer's business operations make it significantly more vulnerable to adverse operating conditions than the IICRA score indicates. 6 (weak) An insurer's business operations make it very considerably more vulnerable to adverse operating conditions than the IICRA score indicates A majority of subfactors is positive and it is rare for one to be negative. Positive subfactors clearly outweigh negative subfactors. Generally, positive subfactors slightly outweigh negative subfactors. Positive and negative subfactors balance each other; or most subfactors are neutral and a minority is negative. Negative subfactors outweigh positive subfactors if any. Most subfactors are negative and it is rare for one to be positive. *The score is no stronger than '5' if either the insurer's gross annual premiums or its total assets do not consistently exceed approximately $50 million or equivalent. If neither geographic diversification nor other diversification is positive, or if operating performance is negative, the score is no stronger than '3'. Table 9 Assessment Of Operating Underperformance And Outperformance Operating performance metrics* All insurers: return on capital Examples of consistent and material underperformance The insurer's average return over the past five years is one-quarter (e.g., 7.5% versus 10%) below the peer group average, or over the past two years is less than half of the peer group average. Examples of consistent and material outperformance The insurer's average return over the past five years is one-quarter more than the peer group average. Life insurers including composite insurers New business margin Return on assets, prebonus pretax earnings/total assets, and return on embedded value P/C insurers including composite insurers Return on revenue and combined ratio The average return over the past five years is 25 basis points lower than the peer group average, or over the past two years is 50 basis points lower than the peer group average. For any of the three metrics, the insurer's average return over the past five years is one-quarter below the peer group average, or over the past two years is less than half of the peer group average. The average return over the past five years is for either metric 5 percentage points lower than the peer group average, or over the past two years is for either metric 10 percentage points weaker than the peer group average. The average return over the past five years is 25 basis points higher than the peer group average. For two of the three metrics, the insurer's average return over the past five years is one-quarter more than the peer group average. The average return over the past five years is for either metric 5 percentage points stronger than the peer group average. *These metrics are defined in "Assumptions For Quantitative Metrics Used In Rating Insurers Globally," published on April 14, Not all of these metrics are applicable or available for all insurers involved in life insurance because of different accounting and reporting frameworks. The assessment is based on those that are applicable and available. These metrics are used only if the new business margin metric is not available. 1. Operating performance 72. The analysis of the operating performance subfactor complements that of the other subfactors in that a "positive" score JULY 9,

22 is a likely consequence of a healthy competitive position. An insurer achieves a "positive" score if its operating performance, as defined in table 9, is materially stronger than that of its peers. 73. It may also achieve a positive score if its gross expense ratio (P/C) or general expense ratio (life) is consistently at least 10% lower than the average of peers with similar distribution channels and similar products (i.e., under 27% if the average is 30%). These ratios are defined in 150 and 113 of "Assumptions For Quantitative Metrics Used In Rating Insurers Globally," April 14, An insurer has a "negative" score if it consistently and materially underperforms peers, as defined in table 9. It has a "neutral" score if it does not meet the requirements for either a "positive" or "negative" score. 2. Differentiation of brand or reputation 75. The insurer's differentiation of brand or reputation relative to its peers is assessed from the perspective of current or potential policyholders and, for intermediated business, their intermediaries. 76. Most insurance markets are competitive and commoditized to a large degree, leaving most industry participants relatively undifferentiated from their peers. Therefore, most insurers are likely to be scored "neutral." Only a small minority of rated insurers are likely to achieve a "positive" assessment. An insurer has a "neutral" score if it does not meet the requirements for either a "positive" or "negative" score. 77. If at least one of the following applies to a very substantial proportion--typically 50% or more of consolidated gross premiums--of the overall business of a insurer, the score is "positive": The insurer has consistently positive media commentary or consistently positively differentiated results in policyholder or intermediary surveys. The insurer is consistently successful in product innovations, i.e., the organization is early to market in new product design, and is among the first to raise premium rates when rates start to rise or among the last to lower premium rates when rates start to fall. A majority of its business is written on a subscription or coinsurance basis (this applies to reinsurance and large commercial or industrial business); and the insurer is a leader in terms of premium rates and product design. Leaders significantly influence premium rates and product design. 78. If at least one of the following applies to a significant proportion (typically 50% or more of consolidated gross premiums) of the overall business of an insurer or reinsurer, the score is "negative": It has consistently negative media commentary or consistently negatively differentiated results in policyholder or intermediary surveys. It is consistently unsuccessful in product innovations, i.e., it is slow to market in new product design, or is among the last to raise premium rates when rates start to rise, or among the first to lower premium rates when rates start to fall. A majority of its business is written on a subscription or coinsurance basis (this applies to reinsurance and large commercial or industrial business); and the insurer is a follower, rather than a leader, in terms of premium rates and product design. JULY 9,

23 3. Market share 79. The market share subfactor is scored either "positive" or "neutral." Market share is measured by an insurer's share of gross premiums (for P/C insurers and for reinsurers) or policyholder obligations (for life insurers) for the market where it participates. 80. The subfactor is scored "positive" if an insurer has a strong market share. Examples of strong market share include, for outperforming insurers (see table 9), the following: The insurer has a sustainable global market share of approximately 20% or more in one of the following global industries: P/C reinsurance, life reinsurance, trade credit insurance, and marine P&I insurance; The insurer has a sustainable market share of approximately 20% or more in at least one significant country or significant U.S. state; or The insurer consistently ranks among the top five insurers by market share, or its market share is at least 90% of that of the fifth-largest, in three or more significant markets. 81. For the purposes of assessing market shares, a significant market is defined as a P/C insurance line of business or a life or health insurance product type for each significant country (see table 10). Table 10 Definition Of Significant Market Sector Significant countries, U.S. regions or states* Lines of business or product types P/C lines of business Significant countries: U.S., Canada, Japan, China, U.K., France, Germany, Italy, Spain, Netherlands, Australia, Brazil, South Korea, and Russia Significant U.S. regions: Northeast, Midwest, West, South Significant U.S. states: California, Florida, New York, and Texas Life/health insurance product types Significant countries: U.S., Canada, Japan, China, U.K., France, Germany, Italy, Spain, Sweden, Switzerland, Netherlands, Australia, Brazil, South Korea, India, Taiwan, and South Africa Significant U.S. regions: Northeast, Midwest, West, South Significant U.S. states: California, Florida, New York, and Texas Auto/motor (liability and property), personal property, commercial property, ships, aircraft and cargo (liability and property), workers' compensation/employers' liability, other liability, personal accident and short term health, and credit/surety/pecuniary Individual life or long-term health protection, group life/health protection, group pension, unit-linked or separate account savings (including U.S. variable annuities), nonunitized savings (including with-profit and U.S. fixed annuities), and annuities (or pensions) in payment *These criteria assumptions reflect sector total insurance premiums written exceeding $30 billion in each country based on Swiss Re's Sigma study "World Insurance in 2011," but excluding financial centers comprising mainly captive insurers of corporates or global insurers. Includes reinsurance of these lines of business. Examples of significant markets would include: German individual life/health insurance protection, reinsurance of Japanese P/C insurance, and insurance of Californian workers compensation. 82. Insurers with large market shares in nonsignificant countries are not scored "positive," nor are insurers with large market shares in narrow subclasses of business (e.g., subclasses of "other liability," see table 10). JULY 9,

24 4. Level of controlled distribution channels 83. The proposed criteria assess the degree to which an insurer can control, or significantly influence, its distribution channels. The assessment factors in affiliates' distribution channels. 84. Controlled distribution channels include: Direct marketing (direct mail, telephone, Internet), Employed sales forces, Distribution through the insurer's affinity groups (including certain industries, professions, associations, trades unions, public service employees, and the armed forces), Distribution through banks not owned by the insurer or its parent or under common control, but under exclusive bancassurance contractual relationships (i.e., the bank's network distributes only that insurer's policies) that we expect will last a decade or more, and Tied agents (see glossary). 85. Controlled distribution channels do not include: Independent intermediaries or brokers, Non-tied agents (see glossary), Premiums produced from price comparison websites, and Distribution through banks not owned by the insurer or its parent, or under common control that does not qualify as controlled under the preceding paragraph. 5. Geographic diversification 86. Diversification, particularly geographic, is fundamental to the insurance business. 87. The subfactor is scored "positive," "neutral," or "negative" based on: The insurer's geographic presence, i.e., the number of those countries with a large surface area and a significant market size, where the insurer writes business; and The level of insurance penetration, defined as in 58, in that geographic area. 88. The score, if not positive or neutral, is negative. Table 11 provides guidance to score this subfactor, subject to the following: In all countries apart from the U.S., a meaningful presence in all significant populated regions of a country is necessary for the insurer's presence to support a "neutral" or "positive" assessment. Insurers primarily present in certain fast-growing markets, including certain large emerging economies, tend to be scored as neutral under this subfactor because the criteria incorporate strong sustainable market growth elsewhere, notably in the IICRA. Table 11 Scoring Geographic Diversification Geographic presence and market size Insurance penetration Score Large Significant Positive Scoring examples: any one of the following (a) to (f) situations applies: JULY 9,

25 Table 11 Scoring Geographic Diversification (cont.) (a) More than 50% of the insurer's business is derived from one or more of the four global insurance sectors defined in 36 to 38. (b) It is globally diverse; that is, it is present in the U.S., Europe, Canada, or developed Asia. For example, it is present in three of these countries or regions where each country or region represents 15% or more of premiums, or is present in at least two of these countries or regions, where each represents 25% or more of premiums. (c) It is diverse within the U.S. For example, it is present in (i) 20 or more U.S. states where each state represents more than 2% of premiums; or (ii) 15 or more U.S. states where each state represents more than 3% of premiums; or (iii) 10 or more U.S. states where each state represents at least 2% of premiums, and where five or more of these states each represents more than 10%. (d) It is diverse within Europe. For example, it is present in (i) 12 or more European countries where each represents more than 4% of premiums; or (ii) nine or more European countries where each represents more than 5% of premiums; or (iii) six or more European countries where each represents at least 4% of premiums, and where three or more each represents more than 10%. (e) It is diverse within developed Asia. For example, it is present in three or more of Japan, China, South Korea, India, Taiwan, and Australia, where each represents more than 15% of premiums. (f) It is diverse within China, using principles similar to those applied for the U.S. in (c) above. Large Less significant Neutral Scoring examples: Insurers in large developing insurance markets, e.g., with operations representing more than 50% of premiums in countries such as China, India, or Brazil. Small Significant Neutral Scoring examples: Insurers (not scored as "positive") in developed insurance markets with operations representing more than 50% of premiums in any of the following: the U.S., Canada, Europe, Japan, Australia, New Zealand, Singapore, South Korea, or Hong Kong. Note: In this table, "Europe" and "European countries" designate countries of the EEA (European Economic Area) and Switzerland. "Developed Asia" designates Japan, China, South Korea, India, Taiwan, and Australia. Premiums designate an insurer's total gross premiums. 6. Other diversification 89. An insurer's competitive position may benefit from other sources of diversification, which is assessed according to table 7. D. Assessing The Financial Risk Profile D1. Deriving The Financial Risk Profile 90. The proposed criteria view the financial risk profile (FRP) as the consequence of decisions that management makes in the context of its business risk profile and its risk tolerances. These decisions include the extent and manner in which the insurer is capitalized, factoring in prospective growth and retained earnings, and the amount and types of financial flexibility it maintains, relative to its risks. 91. The starting point for evaluating an insurer's FRP is the analysis of capital and earnings ( ), including the regulatory capital filter and the representativeness of modeling modifier, resulting in a score on a '1' to '8' scale. We will then adjust this score, as described in table 12 below, for the risk position ( ) and financial flexibility ( ) scores. For example, the combination of a capital and earnings score of '4' with a strong risk position and JULY 9,

26 weak financial flexibility scores would result in an FRP of '5', i.e. (4-1)+2. The FRP score ranges from '1' to '10'. 92. An insurer's FRP score is also capped by its total asset quality: at '3' when the latter is "adequate," at '7' when it is "less than adequate," and at '8' when it is "weak." Total asset quality is the weighted-average credit quality of bonds, loans, and bank deposits representing shareholders' equity and nonparticipating insurance liabilities. It is "adequate" when weighted-average credit quality is in the 'BBB' category, "less than adequate" when in the 'BB' category, and "weak" when 'B+' or lower. However, total asset quality improves by one category if the investment portfolio diversification subfactor in risk position is "positive" (if average credit quality is in the 'BBB' category, the FRP is not capped). D2. Capital And Earnings 93. Capital and earnings measures an insurer's ability to absorb losses by assessing capital adequacy prospectively, using quantitative and qualitative measures. Capital adequacy first compares currently available capital resources with capital requirements by applying Standard & Poor's capital model and then assesses the insurer's ability and willingness to build capital through net retained earnings and thereby fund growth. 94. Under the proposed criteria, an insurer's capital and earnings is a function of three subfactors: regulatory capital adequacy, capital adequacy, and representativeness of modeling. The last two are relevant only when the first one is JULY 9,

27 scored "less than adequate" or "adequate." 95. Capital and earnings is scored on a scale of '1' to '8', where '1' is the strongest. 96. If the representativeness of modeling subfactor is scored "positive," a capital adequacy score of '7' moves to the stronger score of '6' and a score of '8' moves to '7'. 97. If the representativeness of modeling subfactor is scored "negative," a capital adequacy score of '1' moves to the weaker score of '2', a score of '2' moves to '3', and a score of '3' moves to '4'. 1. Regulatory capital adequacy 98. The regulatory capital adequacy subfactor measures near-term regulatory intervention risk, principally through the buffer between the insurer's available regulatory capital (see box 3) and its minimum capital requirement (MCR) that may trigger the regulator's "ultimate regulatory action" to address current or expected deficiencies in capital or liquidity over the next year. Ultimate regulatory action includes required closure to new business, withdrawal of license, or placement under formal regulatory control. 99. If no clear quantitative regulatory capital requirement exists against which to measure intervention risk, the analysis looks for evidence in the regulatory framework and historical experience to gauge the proximity of intervention. Regulatory requirements are clear, for example, in the U.S. and will be in Europe as well under the EU's Solvency II directive. In European countries, under Solvency I, and in most other countries they are currently less clear The subfactor is scored "adequate," "less than adequate," or "weak." It cannot be scored more favorably since the remoteness of regulatory intervention does not enhance creditworthiness If the evidence suggests that the regulator is unlikely to take action, a score of "adequate" will be assigned Where quantitative regulatory capital requirements are clear, the score is "weak" if the insurer's regulatory capital adequacy ratio: At the latest measurement date, was at or below 1.2 times (x) the MCR and is likely to remain so on the next measurement date, or Given observed and potential volatility, is highly likely to breach that 1.2x level at its next measurement date or within one year, whichever is later Where quantitative regulatory capital requirements are clear, the score is "less than adequate" if the insurer's regulatory capital adequacy ratio: At the latest measurement date, was only marginally above the MCR and is likely to remain so at the next measurement date. Typically, "marginally above" would mean within 1.2x-1.5x the MCR under the proposed criteria, but this could vary according to Standard & Poor's expectation of the regulator's behavior. Is likely to be only marginally above the MCR at the next measurement date or within one year, whichever is later The score is "adequate" in all other cases. JULY 9,

28 2. Capital adequacy 105. The capital adequacy subfactor considers capital prospectively by evaluating, at four confidence levels, the amount of TAC relative to risk-based capital requirements that an insurer is likely to hold over the current and next two years to cover, over the expected life of its portfolio, losses from the various risks it carries. Prospective capital adequacy is the level where, at the end of the projection period, TAC most closely matches the RBC requirement (see table 13), subject to earnings quality and to the front- versus back-loaded nature of any improvement in capital adequacy Capital adequacy is scored from '1' to '8', where '1' is strongest The primary source of capital generation depends on an insurer's ability and willingness to increase TAC through retained earnings. Thus, the quantitative analysis compares an insurer's TAC with its RBC requirements at four confidence levels, factoring in prospective TAC generation, RBC requirement growth, and changes in risk profile. Additions or reductions are made to RBC requirements as determined in "A New Level Of Enterprise Risk Management Analysis: Methodology For Assessing Insurers' Economic Capital Models," published on Jan. 24, The size factor adjustment and invested asset concentration risk charge in the capital model criteria (see 127 to 132 of these criteria) do not apply since both are already factored in the proposed representativeness of modeling and investment portfolio diversification subfactors The qualitative analysis is performed through an assessment of earnings quality (see 113 to 114). Earnings quality informs the overall capital and earnings assessment for an insurer by setting higher benchmarks for TAC generation. JULY 9,

29 109. The capital adequacy analysis follows four steps: For the past financial year-end, the insurer's TAC and RBC requirements at various confidence levels are determined through the capital model's quantitative analysis. For the end of the current and two subsequent years, the RBC requirement is projected using the RBC requirement at the end of the previous financial year and the rate of RBC requirement growth or contraction over the year. For example, if the RBC requirement was $12,000 at year-end 2011, and we expected 5% business growth, the RBC requirement at year-end 2012 would be $12,600. Earnings quality is scored as "high" or "low," according to 113 and 114. The impact of earnings quality is described in 115 and 117. Prospective TAC generation and TAC are calculated, as described in 115 to 117, for each of the current and two subsequent years The capital and earnings score is set by comparing, according to table 13, the redundancy or deficiency of TAC relative to the RBC requirement at four confidence levels at the end of the projection period. If earnings quality is low, the projection does not strengthen the score by more than two notches from the level the last financial statements imply; if earnings quality is high, the projection does not strengthen the score by more than three notches. This is to reflect the inherent uncertainties in projecting a sustainable improvement in capital adequacy. In addition, a stronger score based on projections applies only if the improvement occurs primarily in the current and next year For a life insurer, RBC requirement growth is typically the growth rate in total assets, defined as the sum of investment return and net flows, which themselves are new premium and deposits, less claims payments. For a P/C insurer, RBC requirement growth is typically that of net premiums written. For a multiline insurer, RBC requirement growth is first computed separately for its life and P/C businesses, and then summed In addition, the growth assumptions in the previous paragraph incorporate the evaluation of any material changes in the risk profile that could influence future RBC requirements. The effect of foreign exchange is assessed as neutral unless particular foreign exchange mismatches exist for an insurer and the mismatch is in a volatile currency. JULY 9,

30 113. Earnings Quality. Earnings quality is assessed as "low" or "high." "Low" earnings quality indicates higher-risk, more-volatile earnings that add uncertainty to forecasts for an insurer's earnings and capital Earnings quality is scored high if all of the following features are present on a continuing basis, where "modest proportion" typically indicates approximately one-quarter: An insurer's earnings comprise mostly operating earnings, rather than the contribution from realized or unrealized investment gains and losses. Standard & Poor's capital model's one-in-250-year catastrophe charge (see 152 to 155 in the capital model criteria) is less than annual operating earnings. Spread-based earnings are a modest proportion of operating earnings. Losses from a hypothetical 25% decline in equity markets represent at most a modest proportion of operating earnings, whether or not the losses flow through the profit and loss statement. The insurer's premium-weighted IICRA (see 33 to 38) is no weaker than '3'--otherwise the industry and country environment raises uncertainty in forecasts The absolute growth in TAC is calculated as after-tax earnings, minus other changes in TAC. If earnings quality is low, any positive absolute growth in TAC is halved, except for the portion that has already materialized (for example, a completed equity issue) since the most-recent financial data were released. JULY 9,

31 116. "Other changes in TAC" is calculated as: Dividends we expect an insurer to declare. Although high dividends are a buffer against earnings volatility for an insurer with a flexible dividend policy, insurers are generally reluctant to cut dividends, except when under severe stress. Therefore, when planned dividends are higher than the last ones paid, the calculation takes them into account, and when they are lower, the calculation takes the dividend cut into account only when it is reasonably certain. Plus common equity repurchases that we expect the parent company to make. Minus equity issuance that is reasonably certain to happen in the first year. Plus or minus additional changes in TAC, such as changes in the revaluation reserve since the last balance sheet date, or 50% of the change in value of in-force that we expect In the example in table 14, earnings quality is "high" and regulatory capital adequacy "adequate." The insurer's TAC is growing faster than its business, strengthening capital adequacy. At year-end 2011, TAC was closest to a 'BBB' RBC requirement level and would have been scored '5' without forward-looking analysis. But the analysis indicates that TAC will be closest to an 'A' level at year-end 2014, and it is therefore strong enough at year-end 2012 for a capital adequacy score of '4'. Table 14 Illustrative Example Of Scoring Of Capital Adequacy RBC requirement growth (see 117 and 118) during the year Year-end RBC requirement in U.S. dollars at various confidence levels Forecast for subsequent year (e.g. 2014) Forecast for next year (e.g. 2013) Expected for current year (e.g. 2012) End of last year (e.g. Dec. 31, 2011) 5% 5% 5% N/A At 'AAA' $13,892 $13,230 $12,600 $12,000 At 'AA' $12,734 $12,128 $11,550 $11,000 At 'A' $11,576 $11,025 $10,500 $10,000 At 'BBB' $9,261 $8,820 $8,400 $8,000 Beginning-of-year TAC $10,450 $9,900 $8,900 N/A Aftertax operating income as component of change in TAC (after minority interests) during the year Other changes in TAC (see 122) during the year $1,500 $1,200 $1,500 N/A $-700 $-650 $-500 N/A Year-end TAC $11,250 $10,450 $9,900 $8,900 Rating symbols in this table refer to confidence levels in 11 of the capital model criteria. N/A--Not applicable. RBC--Risk-based capital. 3. Representativeness of modeling 118. The representativeness of modeling subfactor determines whether the analysis of prospective capital adequacy has overstated or understated capital and operating performance. This differs from the risk position factor, which identifies unmodeled risks that can affect capital volatility The representativeness subfactor is scored "positive," "negative," or "neutral." Most insurers are likely to be scored "neutral" under the proposed criteria. JULY 9,

32 120. Representativeness of modeling is positive if the capital model materially overstates specific product risks of the insurer Representativeness of modeling is scored "negative" if: The capital model materially understates specific product risks of the insurer; TAC is small (for example, for U.S. insurers, under $1 billion), making it more vulnerable to single-event losses beyond that assumed in the capital model; Acquisition or shareholder-distribution risks may weaken capital adequacy beyond what can be reliably quantified; or The capital model results depend heavily on weaker forms of capital (for example, if value of in-force or other weak forms of capital contribute more than 50% of TAC). D3. Risk Position 122. Risk position assesses material risks that the capital model does not incorporate and specific risks that it captures, but that could make an insurer's capital and related financial ratios significantly more, or significantly less, volatile Risk position is scored on a scale of '1' to '4', where '1' is strongest, based on an analysis of five subfactors (see table 16) The five risk position subfactors are: Exposure to employee benefits (see 125 to 128), Foreign currency exposure (see 129 to 133), Investment leverage (see 134 to 138), Investment portfolio diversification (see 139 to 147), and Additional sources of capital volatility (see 148 to 150). JULY 9,

33 Table 16 Risk Position Assessment Score Descriptor Guidance (see table 15 and 125 to 150) 1 Strong The insurer's prospective capital adequacy has a low volatility risk, and the insurer has no material risk concentrations. This is typically associated with none of the subfactors being scored negative, and one or more being scored positive. JULY 9,

34 Table 16 Risk Position Assessment (cont.) 2 Adequate The insurer's prospective capital adequacy has average volatility risk, or certain risks are not incorporated in the capital model or risk concentrations exist, but are not material. This is typically associated with all subfactors being scored neutral; or with "additional sources of capital volatility" being scored neutral or positive, and at most two of the other subfactors being scored negative. 3 Less than adequate The insurer's prospective capital adequacy has somewhat above-average volatility risk, or certain risks are not incorporated in the capital model or risk concentrations exist, and these may be material. This is typically associated with "additional sources of capital volatility" being scored negative; or with three subfactors being scored negative when the ERM risk controls subfactor is not scored negative. 4 Weak The insurer's prospective capital adequacy has clearly above-average volatility risk, or certain risks are not incorporated in the capital model and risk concentrations exist, and these are significant, or some investment risk characteristics exist that could cause severe capital stress, i.e., the capital position would be so weakened that the company is no longer able to effectively compete in markets, and may simultaneously face liquidity strain. This is typically associated with four or five subfactors being scored negative; or with three subfactors and the ERM risk controls subfactor being scored negative. 1. Exposure to employee benefits 125. This subfactor seeks to capture an insurer's exposure to employee postemployment benefit obligations (including pension and retiree health care benefits) in terms of both liability and asset risks, whether these are recognized on the balance sheet or not. To do this, the proposed criteria calculate the ratio of total pretax gross employee benefit obligations (for pensions, the projected benefit obligation), net of any surplus, to TAC The subfactor is scored "negative" if the ratio indicates that employee benefits represent a high proportion of TAC, typically of more than one-quarter. Otherwise, the subfactor is scored "neutral." 127. If an insurer does not disclose liabilities and asset values with sufficient precision to calculate the ratio, unless there is other evidence to indicate that employee benefit risk is low, the subfactor is scored "negative." 128. This subfactor complements the capital model by reflecting the risk of the net employee benefit obligation changing significantly, as well as the significant dependence of asset and liability values on choices of key assumptions, including sustainable asset values, life expectancy, discount rates, and future increases in pensionable earnings. 2. Foreign currency exposure 129. This subfactor assesses currency mismatches between assets and liabilities, which the capital model does not capture The score is "neutral" unless there is a significant mismatch. In that case, the score is "negative." 131. The proposed criteria define a significant mismatch as a situation where, for one or more material currencies, the mismatch of assets to liabilities in that currency is consistently about 10% or more in either direction Exposure to foreign exchange rate fluctuations can pose a risk to capital for an insurer with material obligations denominated in currencies other than its primary operating currency. Most insurers mitigate this risk by holding foreign currency assets of a similar amount to their foreign currency liabilities. If a company has demonstrated a good track record of effectively hedging currency mismatches, its hedged amounts would not be included when calculating the mismatch. For example, euro-denominated assets swapped to yen in support of yen liabilities would be excluded from the mismatch calculation The criteria focus on material foreign exchange exposure because an insurer operating in numerous foreign JULY 9,

35 jurisdictions will often post mismatches that may not be practical or efficient to eliminate at all times. Such mismatches do not appreciably alter an insurer's capital position. 3. Investment leverage 134. The investment leverage subfactor identifies a very high or low proportion of high-risk assets in the asset base. To do this, the criteria calculate the investment leverage ratio, defined as the ratio of volatile or illiquid assets to TAC. The capital model does not fully capture an insurer's exposure to volatile or illiquid investments. The score is "neutral," "positive," or "negative." 135. Slightly different cutoffs in the criteria apply to an insurer if its liabilities with significant profit-sharing characteristics (such as participating whole-life policies in the U.S. and "with-profit" contracts sold in Europe) exceed approximately one-quarter of general account liabilities. These insurers can transfer a higher proportion of the investment risk associated with high-risk assets to their policyholders by reducing crediting rates, bonuses, or dividends to policyholders For insurers with significant profit sharing, the subfactor is scored positive if the investment leverage ratio is consistently low (for example, under about 20%), negative if significantly above 100% (for example, over 150%); otherwise, the subfactor is scored neutral For other insurers, which cannot transfer as much of the investment risk to policyholders, the subfactor is scored "positive" if the investment leverage ratio is consistently very low (for example, under about 10%), and "negative" if consistently in excess of 100% For the purpose of this subfactor, the criteria define high-risk assets as the sum of the values, net of hedges, of: Speculative-grade or unrated bonds, loans, and deposits at speculative-grade banks; Unaffiliated equity investments; Equity real estate assets, excluding those the insurer uses for its own operations and those in markets that the capital model identifies as lower risk, like Switzerland; and Investments in partnerships, joint ventures, and other alternative investments. 4. Investment portfolio diversification 139. The investment portfolio diversification subfactor addresses the risk of an insurer's exposure to a given asset sector or obligor. The capital model does not factor in correlation risks beyond an assumed average degree of asset diversity. This subfactor identifies insurers with more or less risk than the average by using ratios that gauge sector and obligor concentration The subfactor is scored "positive" when the investment portfolio is well diversified among sectors and obligors, that is, typically where no more than 15% of the portfolio is held within any one sector and no more than approximately 5% per obligor The subfactor is scored "neutral" when the investment portfolio is moderately diversified among sectors and obligors--when concentrations do not exceed 15%-30% to any one sector, or 5%-10% to any one obligor The subfactor is scored "negative" in all other cases. JULY 9,

36 143. The sector ratio in 140 and 141 is, for each asset sector (as defined in 147) in an insurer's portfolio, the ratio of the related asset value (combining all obligations related to that asset sector) to the insurer's total invested assets The obligor ratio in 140 and 141 is, for each obligor in an insurer's portfolio, the ratio of the insurer's investment in that obligor's equity and debt obligations to total invested assets For both ratios, the proposed criteria define total invested assets as the sum of cash, cash equivalents, and invested assets, excluding government obligations. These government obligations are also excluded from the numerator of the sector and counterparty ratio tests. This is because the proposed criteria already capture the related risk, notably in the sovereign test in section E4. "Government obligations" are defined as financial obligations issued or guaranteed by an insurer's domestic national government, or those of domestic GREs, including banks, with an "almost certain" or "extremely high" likelihood of extraordinary government support (according to 18 and table 1 of "Rating Government-Related Entities: Methodology And Assumptions," Dec. 9, 2010) The proposed criteria define "domestic national government" as the sovereign governments where the insurer conducts material operations, or those in whose currency the insurer has underwritten policies. In the latter case, given the absence of currency mismatch, the obligations are excluded only up to the extent they cover such insurance policies For the purposes of the above-mentioned ratios, the following definitions of "sector" apply: For U.S. municipal bonds: tax-backed and appropriation-backed government obligations, municipal water sewer obligations, and public university obligations are aggregated by state and each state is viewed as a sector. In addition, the following types of municipal bonds are viewed as individual sectors on a national basis: private education, health care, housing revenue, transportation, public power and other utilities, and other not-for-profit obligations. For structured finance securities, each of the following is defined as a sector: residential mortgage-backed securities (RMBS); commercial receivables; autos; credit cards; student loans; commercial real estate (CRE), including CRE CDOs (collateralized debt obligations); CDOs of asset-backed securities (ABS); all else, including corporate CDOs. For corporate securities, equity and fixed-income obligations are aggregated per issuer, and issuers are allocated to the following sectors: financial services (excluding covered bonds), industrials, technology, transportation, real estate, consumer products, retail and restaurant, energy, and utilities. 5. Additional sources of capital volatility 148. The proposed criteria broadly analyze the volatility of capital across all insurance sectors and consider the unique risks that each insurer has within its exposure portfolio. The assessment takes into account several sources of volatility. Capital stress, relative to capital model results, can be associated with either model error or event risks that, although remote, are not otherwise captured in the risk position analysis. Capital stress factors also include prospective macroeconomic trends such as pronounced inflation or deflation, insured claims trends, and accounting changes Additional sources of capital volatility are scored "positive" when the volatility in the capital model outcomes is low for an insurer, "negative" when it may be high, and "neutral" otherwise Further examples include deficiency in reinsurance protection corresponding to the risk profile or unprotected natural catastrophe risk, terrorism risk, and extreme mortality or morbidity risk. For life insurers that issue variable annuities JULY 9,

37 with guaranteed living benefits, unhedged market exposures can be an acute source of capital deterioration in stressed market conditions. D4. Financial Flexibility 151. Financial flexibility uses qualitative and quantitative measures to estimate the balance between an insurer's sources and uses of external capital and liquidity over the current and next two years The assessment focuses on external sources of capital and liquidity because the proposed criteria assess internal sources through other factors As shown in table 17, the criteria use three subfactors to assess financial flexibility. Each is scored "positive," "neutral," or "negative": Access to sources of external capital and liquidity (see 158 to 169), Financial leverage (see 170 to 176), and Fixed-charge coverage (see 177 and 178). Table 17 Assessing The Subfactors Of Financial Flexibility Subfactors Positive Neutral Negative Access to sources of external capital and liquidity (see 158 to 169) Financial leverage (see 170 to 176) Fixed-charge coverage (see 177 and 178) Access to at least three sources, each substantial with market access that significantly exceeds the insurer's liquidity and capital needs Access to several sources with sufficient available market access that exceeds the insurer's liquidity and capital needs Access to limited sources or access to sources that have only limited or inadequate available market access relative to current or future needs Financial leverage is low Financial leverage is moderate Financial leverage is high Fixed-charge coverage is high Fixed-charge coverage is moderate Fixed-charge coverage is low 154. The last two subfactors consider that a company with high leverage and a low fixed-charge coverage ratio is likely to have less capacity and flexibility to attract external capital The scores on the subfactors combine to determine the assessment of an insurer's financial flexibility, on a scale of '1' ("strong") to '4' ("weak") according to table Under the proposed criteria, the financial flexibility metrics are evaluated at the consolidated group level for the GCP and for the SACP of group members assigned "core," "highly strategic" or "strategically important" group status. For other group members, the financial flexibility metrics are evaluated at the subsidiary level; a subsidiary's financial flexibility score is typically capped by its parent group's financial flexibility. This is because a subsidiary's access to the market would likely be constrained if it belonged to a weaker group However, a subsidiary's financial flexibility could exceed its majority owner's financial flexibility if a substantial minority interest in it is held by a stronger group or held publicly and widely. JULY 9,

38 Table 18 Financial Flexibility Assessment Score Descriptor Guidance (see table 17) 1 Strong All subfactors are positive. 2 Adequate All situations not covered by the three other descriptors. 3 Less than adequate The "access" subfactor is negative and neither of the other two are negative. 4 Weak The "access" subfactor is negative and one or more of the other subfactors are negative. 1. Access to sources of external capital and liquidity 158. The first financial flexibility subfactor addresses the insurer's ability to access sufficient amounts of external capital or liquidity. The subfactor assesses two main sources of capital or liquidity, as well as other sources of financial flexibility. The assessment is informed by five secondary metrics The subfactor is scored by assessing two main sources of capital or liquidity: The diversity in accessible capital markets (such as public markets for common equity, debt and hybrid instruments, and commercial paper), as shown by the insurer's history of market access; and Holdings of assets with significant unrealized capital gains that could be sold to enhance liquidity without adversely affecting asset-liability matching or future earnings prospects and are not recognized in reported capital because of an accounting framework that uses historical cost and not fair market value Other sources of financial flexibility include: An insurer's ability or demonstrated willingness to obtain reinsurance from reinsurers with higher credit quality than the insurer; An insurer's ability to use securitization techniques to source capital and liquidity; For a life insurer, the ability to adjust policyholder bonuses-dividends or crediting rates to an extent not recognized in the capital and earnings factor; and For certain P/C insurers, such as marine P&I insurers, the ability to retroactively raise premiums The assessment of access to sources of external capital and liquidity considers five secondary metrics (see 164 to 169) relating to: market indicators, capital-raising track record, debt leverage, debt maturity profile, and EBITDA interest coverage The subfactor is scored "positive" if the insurer can fully meet its future additional capital and liquidity needs from each of at least three alternative sources The subfactor is scored "negative" if the insurer has limited sources of additional capital and liquidity, or sources with limited or inadequate available market access available relative to needs. In all other cases, this metric is assessed as "neutral." 164. Secondary considerations informing "Access to sources of external capital and liquidity."market indicators include share, hybrid, and debt prices, and observable credit spreads. Trends in an insurer's share price or credit spreads are useful indicators of market sentiment to inform the assessment because they provide insights into the insurer's cost of capital and ability to access cost-effective funding. Favorable trends in the market price of an issuer's equity and moderate or narrowing credit spreads (relative to peers and in absolute terms) are supportive of the subfactor. JULY 9,

39 Materially adverse trends in the market price of an issuer's equity and wide or widening spreads (relative to peers) weigh on the subfactor An insurer's track record in accessing the equity and debt capital markets or regularly receiving parent funding, utilizing reinsurance or securitization, adjusting policyholder bonuses or dividends or crediting rates, or calling for additional premiums on insurance policies already in force provides tangible evidence to support the assessment of its ability to access capital and liquidity. It is also important to consider both its willingness and capacity to access capital and liquidity. Even with an excellent capital-raising track record, limited current capacity drives a negative assessment The debt leverage metric indicates the insurer's capacity to issue additional debt (for the definition, see 94 of "Assumptions For Quantitative Metrics Used In Rating Insurers Globally," April 14, 2011). It is adjusted for deficits on employee postemployment obligations For insurers operating in developed markets, debt leverage of less than 15% supports the insurer's ability to access capital and liquidity. Debt leverage approaching or exceeding 30% is detrimental. Otherwise, the criteria consider it neutral The interest coverage metric provides an indication of the insurer's capacity to issue additional debt based on the strength of earnings available to service its debt obligations (for the definition see 84 of "Assumptions For Quantitative Metrics Used In Rating Insurers Globally," April 14, 2011) For insurers operating in developed markets, interest coverage supports financial flexibility when consistently exceeding 10x and is detrimental when consistently less than 6x. Otherwise, the criteria consider it neutral. 2. Financial leverage 170. The financial leverage subfactor addresses the degree of an insurer's indebtedness relative to its total capitalization. It is proposed as a key metric, instead of debt leverage, because it includes hybrid instruments and is therefore a broader measure of balance sheet leverage, and because it is more comparable globally. Regulatory differences across markets encourage different capital structures that may distort comparisons of debt leverage Financial leverage is measured as defined in 96 of "Assumptions For Quantitative Metrics Used In Rating Insurers Globally," April 14, 2011, and scored "positive," "neutral," or "negative." JULY 9,

40 172. A financial leverage ratio consistently less than 20% is scored "positive" for an insurer operating primarily in developed insurance markets A financial leverage ratio consistently more than 40% is typically scored "negative" for an insurer operating primarily in developed insurance markets. If it is 40%-50% and if debt leverage is less than 25%, the subfactor is scored "neutral." If financial leverage is 20%-40%, the subfactor is also scored "neutral." 174. The ratio cutoff points in the two previous paragraphs are reduced (indicating a more conservative leverage) to reflect an insurer's unfavorable debt maturity profile or ratio of intangibles to equity (see next two paragraphs), two features the financial leverage ratio does not address. An unfavorable debt maturity profile and an unfavorable ratio of intangibles to equity each reduce the ratios by approximately 5 percentage points. A very unfavorable ratio of intangibles to equity reduces the cutoff points by approximately 10 percentage points. For example, the score is "positive," despite an unfavorable debt maturity profile and a 70% ratio of intangibles to equity if financial leverage is less than 10% (20% minus 5 points minus 5 points) The debt maturity profile is unfavorable if several significant maturities are concentrated in the near to medium term. In this analysis, debt maturities include hybrid securities with simultaneous call and step-ups, because we typically expect the issuer to then call the instruments. Unless unfavorable, the debt maturity profile is neutral The ratio of intangibles to shareholders' equity (intangibles ratio) addresses the quality of an insurer's equity. Intangibles exceeding half of equity are typically consistent with an unfavorable assessment, and intangibles exceeding equity are typically consistent with a very unfavorable assessment. For the purpose of this subfactor, intangibles are defined as the sum of goodwill, intangible assets, deferred acquisition costs (DAC), value of in-force, value of business acquired, and deferred tax assets (as reported on the primary and any supplementary financial statements used to calculate financial leverage). Unless unfavorable or very unfavorable, the intangibles ratio is neutral. JULY 9,

41 3. Fixed-charge coverage 177. The fixed-charge coverage subfactor addresses an insurer's ability to service interest on financial obligations out of EBITDA. The proposed criteria view it as a key metric, replacing interest coverage, for the same reasons as the criteria prefer financial leverage over debt leverage. Fixed-charge cover is as defined by 88 of "Assumptions For Quantitative Metrics Used In Rating Insurers Globally," April 14, A fixed-charge coverage consistently exceeding 8x is scored "positive." A fixed-charge coverage unlikely to consistently exceed 4x is scored "negative." Otherwise, the score is "neutral." E. Modifiers And Caps To The Indicative SACP Or GCP 179. The proposed criteria establish four additional factors that are not specific to the business risk or financial risk profiles: The insurer's ERM and management score modifies or caps the anchor according to table 20 to produce the indicative SACP or GCP (see section E1). The insurer's liquidity score is scored from '1' ("exceptional") to '5' ("weak"). A score of '1', '2,' or '3' does not affect the SACP, GCP, or ratings, but a score of '4' caps each at 'bb+' or 'BB+', and a score of '5' caps each at 'b-' or 'B-' (see section E2). The insurer's fixed-charge coverage can constrain the SACP, GCP, or ratings, and is scored according to table 26 (see section E3). The relevant sovereign ratings and T&C assessments are usually a cap, unless an insurer meets certain conditions (see section E4) These caps and modifiers apply cumulatively. For example, an 'aa' anchor combines with a '3' ERM and management score to produce an 'a+' indicative GCP. The group's core operating companies are then rated 'A+' if the three other tests in this section are met. Conversely, the ratings are capped at 'A' if the fixed-charge cover is only 2.5x; 'BB+' if, in addition, liquidity is scored "less than adequate;" and 'BB' if a 'BB' sovereign cap applies Liquidity falls outside the analysis of the business risk and financial risk profiles. An insurer with weak liquidity is considered more likely to default than an identical insurer with stronger liquidity. Even if an insurer's creditworthiness is otherwise consistent with an investment-grade rating, it may fall abruptly under extreme stress if liquidity is "less than adequate" or "weak." This could compromise the insurer's ability to pay claims, despite its other strengths. For that reason, the assessment is absolute and not relative to peers or insurers in the same rating category On the other hand, the proposed criteria do not uplift the SACP or GCP for strong liquidity because it does not enhance an insurer's overall creditworthiness. The criteria assess other rating factors, which may support strong liquidity, when establishing the anchor. E1. ERM And Management Score 183. The proposed criteria combine the two categories of ERM and management and corporate strategy into a single score from '1' to '5', from "very strong" to "weak." The "importance of ERM" is also scored as "high" or "low," depending on whether an insurer is exposed to complex risks that could cause it significant loss of capital or earnings in a short JULY 9,

42 period of time (such as catastrophe risk or high-risk investment exposures) or one that is highly uncertain and usually long term in nature (such as long-tail casualty lines of business). Where the importance of ERM is considered "low," the management and corporate strategy subfactor is the primary factor The ERM subfactor examines whether risk management practices are executed in a systematic, consistent, and strategic manner that provides for the control of losses within an optimal risk-reward framework. ERM analysis allows for a prospective view of the insurer's risk profile and capital needs The management and corporate strategy subfactor addresses how management's strategic competence, operational effectiveness, financial management, and governance practices shape the insurer's competitiveness in the marketplace, the strength of its financial risk management and the robustness of its governance. Stronger management of important strategic and financial risks may enhance creditworthiness. Weak management, with a flawed operating strategy or an inability to execute its business plan effectively, substantially increases an insurer's credit risk (see "Request For Comment: Management And Governance Credit Factors," published March 12, 2012). Table 20 Indicative SACP Or GCP Assessment ERM and management score (from table 21) Anchor (from table 1) aa+ aaa aa+ aa- a+ bb aa aa+ aa a+ a bb aa- aa aa- a+ a bb a+ aa- a+ a+ a- bb a a+ a a a- bb a- a a- a- bbb+ bb bbb+ a- bbb+ bbb+ bbb bb bbb bbb+ bbb bbb bbb- bb bbb- bbb bbb- bbb- bb+ bb bb+ bbb- bb+ bb+ bb b+ bb bb+ bb bb bb- b bb- bb bb- bb- b+ b b+ bb- b+ b+ b b- b b+ b b b- b- b- b b- b- b- b- Note: The indicative SACP or GCP is also subject to 21. ERM--Enterprise risk management. GCP--Group credit profile. SACP--Stand-alone credit profile. JULY 9,

43 1. Enterprise risk management 186. The proposed criteria score an insurer's enterprise risk management on a scale of '1' to '5', from "very strong" to "weak," (see table 22) as the result of the assessment of five subfactors: Risk management culture, Risk controls, Emerging risk management, Risk models, and Strategic risk management All else being equal, an insurer with a higher ERM score, for example one that operates a more-advanced ERM framework effectively is less likely to experience losses outside its predetermined risk tolerances. These are assessed by the management and corporate strategy factor. The group's ERM score is assigned to group members that are "core," "highly strategic," or "strategically important" and are well integrated into the group's ERM processes, such that their processes are virtually indistinguishable. For all other group members, ERM is scored from a stand-alone perspective. Table 22 ERM Assessment Score Assessment Guidance 1 Very strong All subfactors are scored positive. 2 Strong The risk management culture, risk controls, and strategic risk management subfactors are scored positive; one or both of the other two subfactors is scored neutral; and no subfactor is scored negative. 3 Adequate with strong risk controls The risk controls subfactor is scored positive, risk management culture or strategic risk management is scored neutral, and no subfactor is scored negative. 4 Adequate The risk controls subfactor is scored neutral, and the risk management culture and strategic risk management subfactors are both scored positive or neutral. 5 Weak The risk controls or risk management culture subfactor is scored negative. ERM--Enterprise risk management. JULY 9,

44 Table 23 Proposed Scoring Metrics For Enterprise Risk Management (ERM) Risk management culture (see 188 to 190) Risk controls (see 191 and 192) Emerging risk management (see 193 and 194) Risk models (see 195 and 196) Strategic risk management (see 197 and 198) ECM--Economic capital management. Positive Neutral Negative ERM is well established in the organisation with a strong governance structure that addresses all key risks. There is a clear vision of the enterprise's risk profile. Risk limits are aligned with overall risk tolerances. The risk tolerance adopted by the board is clearly communicated internally and externally, and is translated into risk limits. Risks are clearly identified from multiple sources with multiple metrics. Effective processes exist for frequent monitoring, limit enforcement, and clear remediation action steps. Analysis demonstrates a consistent pattern of effectiveness. The insurer has well-established processes with a robust track record of anticipating emerging risks, envisioning their significance and preparing for or mitigating them. The insurer demonstrates an established use of economic capital models, demonstrating an enterprise view. Risk models are used extensively for individual risks across the enterprise and have evolved over time responding to new research and technological capabilities, resulting in state of the art models. Comprehensive validation and sensitivity analyses of all models take place. The insurer optimizes risk-adjusted returns across the enterprise and all of its material risks using economic value metrics, consistently applied and with a good track record. ECM significantly influences capital management and allocation, pricing, performance management, management incentives, and external reporting. Risk tolerances are less clearly defined and communicated. Risk limits are simplistic or do not align with overall risk tolerances. The governance structure covers most key risks but may have limited or infrequent access to the board. Key risks are managed on an individual basis but may lack enterprise oversight. Not all significant risk exposures are identified. There is a limited metric framework for measuring control effectiveness. The framework is generally effective, but limit enforcement policy is less formal or inconsistent than for insurers scored positive. Some processes exist for anticipating emerging risks and envisioning their significance. Generally ERM is limited to identification, lacking processes for mitigation. Risk models are fairly sophisticated and cover most material risks. Comprehensive validation and sensitivity analyses of all models take place. Capital management uses an allocated version of the regulatory or rating agency capital. The insurer uses generic capital formulas instead of a full ECM that reflects an insurer's risks; or the insurer has a limited track record in applying ECM across the enterprise. ERM is not practiced, or is practiced inconsistently, across the enterprise. Risk limits and monitoring do not exist or are very basic and lack alignment with overall risk tolerances. Key risk exposures are not consistently identified or monitored. There is no metric framework for measuring effectiveness or it is not generally consistent; there is no policy for limit enforcement. The insurer lacks processes for evaluating emerging risks. The insurer makes little use of risk models or where it uses them, applies limited validation and sensitivity analysis. The insurer does not optimize of risk-adjusted returns. Capital management activity is only concentrated on maintaining capital levels that are acceptable on a regulatory basis or rating agency's capital adequacy basis Risk management culture. The risk management culture subfactor addresses the importance accorded to risk and ERM in all key aspects of an insurer's corporate decision-making. The analysis assesses the insurer's attitude toward risk, especially its risk appetite framework, risk governance and organizational structure, risk communications and reporting, and the embedding of risk metrics in its compensation structure Supporting evidence typically includes elements such as the presence of an independent ERM function led by an experienced executive, commonly a chief risk officer; a risk profile that is thoroughly understood and well-communicated through a clear risk appetite framework and extensive risk reporting; public disclosures on ERM practices; and a clear linkage between managers' incentive compensation and ERM metrics The risk appetite framework is one of the key elements in assessing an insurer's risk management culture. A "positive" score is assigned where there is a robust ERM framework with a well-defined risk appetite, supported by strong buy-in from the board of directors and business units. An effective risk appetite framework consists of risk preferences and quantitative risk tolerances that have been translated into risk limits, cascaded down to the individual risk and JULY 9,

45 line-of-business levels. A "neutral" score reflects that risks may be managed well individually, but not in a comprehensive and coordinated fashion. A "negative" score implies that ERM is not generally practiced or is not practiced consistently across the enterprise Risk controls. The second subfactor addresses the risk controls an insurer uses to manage key risks, based on its business and risk profile. Such risks typically include credit and counterparty risk, market risk, interest rate risk, insurance risk (including reserving risk), and operational risk. However, the analysis may extend beyond these broad categories (for example, to merger and acquisition risks for an acquisitive insurer) For each major risk category, the analysis considers components of risk control processes such as quality of risk identification, risk monitoring, standards and limits for retained risks, procedures to manage risks within limits, and the effectiveness of remediation when limits are breached. A "positive" score reflects comprehensiveness and a documented history of effectiveness in each of these areas, while a "neutral" score reflects more simplistic methods. These may lack consistency with enterprise risk tolerances, although the controls could still be effective for key risks. A "negative" score indicates that limited risk monitoring is performed or significant weaknesses in effectiveness are noted Emerging risk management. Emerging risk management analyzes how the insurer addresses risks that are not a current threat to creditworthiness, but could become a threat in the future. In addition, it assesses the company's level of preparedness if those emerging risks materialize. Such risks could derive from areas such as regulation, the physical environment, the macroeconomic environment, and medical developments The subfactor is scored "positive" if specific evidence such as internal reports and documentation shows that an insurer has well-established processes to consistently identify, assess, monitor, and potentially mitigate identified emerging risks. A "neutral" score indicates that some processes exist, but the insurer may lack processes for mitigation or measurement. A lack of a formal process to identify emerging risks would result in a "negative" score Risk models. The risk models subfactor assesses the robustness, consistency, and completeness of an insurer's risk model, including, where relevant, its development and use of an economic capital model, and the processes for model governance and validation. Additionally, the subfactor score reflects the risk measures adopted, the methodology, data, and assumptions; the incorporation of risk mitigation activities in those models; the infrastructure to support them; and evidence that their results and limitations are well-communicated and understood by the risk managers and senior management. The analysis is more qualitative than the criteria for assessing an insurer's economic capital model (see "A New Level Of Enterprise Risk Management Analysis: Methodology For Assessing Insurers' Economic Capital Models," published on Jan. 24, 2011) The subfactor is scored "positive" if the insurer's risk model system can perform comprehensive stochastic analysis and deterministic stress scenario analysis, and model results are used in guiding risk decisions. A "neutral" score would indicate effective use of models for material risks, but less comprehensive or robust use in risk decisions. A "negative" score would indicate limited use of models or limited model validation Strategic risk management. The strategic risk management subfactor assesses an insurer's rationale and processes for optimizing risk-adjusted returns, and for evaluating and prioritizing strategic options to accomplish this. The subfactor considers evidence and examples of situations where the insurer has made strategic decisions using economic risk-reward metrics that are consistent with its risk appetite, while balancing other concerns, including regulatory and accounting considerations. JULY 9,

46 198. The score is "positive" if the company demonstrates effective use of risk-reward metrics in strategic planning, product pricing and repricing, its reinsurance strategy, new strategic initiatives (including mergers and acquisitions, entry into new markets, etc.), capital or economic capital budgeting, optimization of risk-adjusted returns, and incentive compensation. A "neutral" score indicates a more-basic approach to risk-reward optimization, based on more-simplistic capital metrics or use of an economic capital model (ECM) with a more limited track record. The score is "negative" where capital management is very basic and no clear risk-reward optimization approach exists. 2. Management and corporate strategy 199. This RFC is consistent with the proposed criteria in "Request For Comment: Management And Governance Credit Factors," published March 12, E2. Liquidity 200. The liquidity analysis centers on an insurer's ability to cover its liquidity needs, both on an ongoing basis and in moderately stressful market and economic conditions. The analysis is absolute, rather than relative to peers The proposed criteria assess an insurer's liquidity on a scale of '1' to '5', where '1' is the strongest, according to table 25. "Adequate" liquidity is rating-neutral at all rating levels; the insurer is able and prepared to cover its liquidity needs so as to survive under moderately stressful conditions for 12 months without any refinancing. "Strong" and "exceptional" liquidity, by definition, exceed the norm. The benchmarks to achieve such levels are correspondingly higher and suggest an ability to weather more stressful scenarios An insurer's liquidity score results from the assessment of four subfactors, each scored as "positive," "neutral," or "negative," according to table 24: Coverage of the insurer's confidence-sensitive liabilities; The possibility that the insurer would need to post collateral; The implications of covenants and ratings triggers in the insurer's financial arrangements; and The insurer's liquidity ratio. Table 24 Liquidity Subfactor Scoring Subfactor Positive Neutral Negative Coverage of confidence-sensitive liability ( 206 to 210) Collateral posting risk ( 211 and 212) Confidence-sensitive liabilities are either non-existent or covered 1.2x or more by the sum of liquid assets and back-up credit facilities not terminated by a one rating category downgrade. Notional exposure to insurance or other contracts where collateral posting is contingently required, assuming a two rating category downgrade or other equivalent triggers, is nonexistent or less than 15% of liquid assets. Confidence-sensitive liabilities are covered approximately 1.2x, or more, by the sum of liquid assets and back-up credit facilities not terminated by a two rating category downgrade. Notional exposure to insurance or other contracts where collateral posting is contingently required, assuming a two rating category downgrade or other equivalent triggers, is within 15%-30% of liquid assets. Confidence-sensitive liabilities are covered clearly less than 1.2x by the sum of liquid assets and back-up credit facilities not terminated by a two rating category downgrade. Notional exposure to insurance or other contracts where collateral posting is contingently required, assuming a two rating category downgrade or other equivalent triggers, is more than 30% of liquid assets. JULY 9,

47 Table 24 Liquidity Subfactor Scoring Covenants and ratings triggers ( 213 to 216) Liquidity ratio ( 217 to 223) (cont.) No financial-ratio covenant nor rating triggers on material facilities. Or, if any, balance sheet and income statement ratios are in excess of 2.0 times (x) covenant requirements. And no material adverse change (MAC) clauses or rating triggers that could result in cancellation of existing facilities. Not at risk to non-economic deterioration in financial metrics linked to covenants due to accounting requirements. The insurer's liquidity ratio exceeds 2.2x. On material facilities, balance sheet ratios are 1.2x-2.0x, and income statement ratios 1.5x-2.0x in excess of covenant requirements. No material adverse change (MAC) clauses or rating triggers that could result in cancellation of existing facilities. Not at risk to non-economic deterioration in financial metrics linked to covenants due to accounting requirements. The insurer's liquidity ratio stands between 1x (P/C) or 1.4x (life) and 2.2x (for both). On material facilities, at least one balance sheet ratios is less than 1.2x, or at least one income statement ratio is less than 1.5x, in excess of covenant requirements. Covenants or triggers are present that if violated would result in liquidity strain or cancellation of existing facilities. Potentially at risk to non-economic deterioration in financial metrics linked to covenants due to accounting requirements. The insurer's liquidity ratio is less than 1.0x (P/C) or 1.4x (life). All assessments in this table are forward-looking over the next 12 months. Table 25 Liquidity Assessment Score Assessment Guidance (see table 24) 1 Exceptional All four subfactors are positive or three are, and the fourth one is neutral.* 2 Strong Two subfactors are positive and two are neutral.* 3 Adequate All four subfactors are neutral or three are, and the fourth one is positive.* 4 Less than adequate One or two subfactors are negative.* 5 Weak Three or all four subfactors are negative or any one subfactor poses a severe risk to the insurer's liquidity. *And no subfactor poses a severe risk to the insurer. For the purpose of the liquidity test, "severe risk" designates an appreciable likelihood that, incorporating a significant but not extreme downside to the insurer's performance expectations under Standard & Poor's base case, one of the four factors renders the issuer unable to entirely and timely service all its financial and policyholder obligations over the next 12 months. Examples include: debt maturities over the next 12 months that are difficult to refinance due to stressed market conditions; and substantial collateral posting requirements related to credit default swaps or other derivative contracts When assessing the liquidity factor to assign a GCP, the analysis is based on a consolidated view including the holding company (to assess the liquidity of a nonoperating holding company, see 249 to 252). Where consolidated data are not available, the analysis includes aggregating operating companies. This analysis excludes insulated subsidiaries. When assessing the liquidity factor to assign an SACP to a specific insurance company, the analysis is restricted to that company, including its subsidiaries, if any. Within a given insurance group, the liquidity scores for the GCP and the various SACPs are capped at "adequate" when the holding company's liquidity is "less than adequate," and at "less than adequate" when it is "weak." 204. The liquidity subfactors consider regulatory or other provisions that may restrict the flow of cash and liquid assets among legal entities, up to the holding company as well as among operating companies, within the rated group. For example, U.S. regulation restricts stockholder dividends by U.S. insurance operating companies. We consider the effect such restrictions may have on a group's ability to meet its liquidity needs in the specific legal entities where they arise. Under the U.S. rules, cash and liquid assets in an insurance operating company may not be available to its holding company to pay maturing commercial paper or to meet the obligations of other insurance operating companies within the group. Where such limitations exist, the subfactors consider only the sources of liquidity available to the legal JULY 9,

48 entities The analysis is based on the following liquidity assumptions and considerations: The ratio cutoffs reflect the historical experience of insurers' liquidity stresses as well as the factors mentioned in 11 to 17. An insurer experiences immediate and unforeseen stress from withdrawals and surrenders over the next 12 months. Refinancing is unavailable for 12 months, i.e., the insurer is assumed not to have any access to market or bank debt, equity, or hybrid instrument refinancing over this period. The insurer's maturities beyond 12 months are manageable; if they are not, the liquidity score is capped at "adequate." For the definition of "liquid assets," see 79 in "Assumptions For Quantitative Metrics Used In Rating Insurers Globally," April 14, 2011, except for the purpose of applying the U.S. and Canada life operating company liquidity model. Backup facilities include only committed credit facilities for general financing with a maturity sufficient to cover liquidity needs (e.g., for liquidity requirements arising in the next 12 months, the credit facilities do not mature within 12 months) and only those provided by banks rated investment grade. When analyzing requirements, including amounts drawn, the entire size of the facility is included as a resource. Alternatively, the analysis can ignore the amounts drawn, but will then consider as a liquidity resource only the facility's undrawn amount. The analysis of an insurer's exposure to rating triggers, collateral posting, and ratio-driven covenants is restricted to instruments and facilities that are material and represent borrowings to third parties, not group affiliates. If not material, instruments and facilities do not contribute meaningfully to liquidity resources. If drawn and accelerated, they can easily be repaid to avoid cross-defaults with larger instruments and facilities. The level of ratio-based covenants is that calculated from the insurer's most recent financial statements. 1. Confidence-sensitive liability coverage 206. The first liquidity subfactor assesses the risk of a sudden call on the insurer's cash in the event of a loss of confidence specific to the insurer or a general loss of market confidence The subfactor is scored "positive" when the insurer has no or insignificant confidence-sensitive liability. It is also positive when the sum of liquid assets and of backup credit facilities not subject to termination in the event of a three-notch downgrade is at least 120% of confidence-sensitive liabilities The subfactor, when not scored positive, is scored "neutral" when the sum of liquid assets and backup credit facilities not subject to termination in the event of a six-notch downgrade is at least 120% of confidence-sensitive liabilities The subfactor is scored "negative" in all other cases Confidence-sensitive liabilities include commercial paper issuance; long-term financial obligations maturing within 12 months; hybrid instruments callable within 12 months; and institutional insurance products, including funding agreements and guaranteed investment contracts (GICs), containing put provisions of up to 365 days or less. 2. Collateral-posting risk 211. The second liquidity subfactor assesses an insurer's exposure, for example through certain debt contracts, reinsurance treaties, and derivatives contracts, to collateral posting requirements in the event of ratings downgrades or other triggers, relative to liquid assets. JULY 9,

49 212. The subfactor is scored "positive" when an insurer has no significant confidence-sensitive liability. It is also positive when the value of the assets it would have to post as collateral in the event of a downgrade of up to six notches or equivalent triggers is less than 15% of liquid assets. The subfactor is scored "neutral" when it would have to post 15%-30% of liquid assets, and "negative" if it would have to post 30% or more. 3. Covenants and ratings triggers 213. The third liquidity subfactor assesses the risk to an insurer's liquidity of not complying with covenants and rating triggers on its third-party financial obligations. The impact of an ensuing termination of debt or credit facilities is a function of the likelihood that the creditor effectively terminates the facility and of the severity of the resulting impact on the insurer's liquidity needs and resources. Examples of such covenants include minimum levels of shareholder equity or statutory capital, or interest coverage by earnings. "Covenant requirement" refers to the most stringent level that, if breached, is defined as an event of default under the documentation The subfactor is scored "positive" for insurers that meet the same requirements as in the previous paragraph, except that balance sheet and income statement numerical measures exceed 2x covenant requirements The subfactor is scored "positive" for an insurer that has no numerical covenant or rating trigger. If it does, the subfactor is "neutral" if the insurer meets all of the following three conditions: The insurer's income statement measures are between 1.5x and 2.0x covenant requirements, and balance sheet measures are between 1.2x and 2.0x covenant requirements. There are no particularly broadly or loosely worded material adverse change (MAC) clauses, or rating triggers within six notches of the current rating that could result in the cancellation of sizable facilities. The insurer is not at risk, over the next 12 months, of a breach of covenants because of a noneconomic deterioration in financial metrics primarily triggered by a change in accounting standards The subfactor is scored "negative" if any of the following applies: One or more income statement measures is less than 1.5x covenant requirements; One or more balance sheet measures is less than 1.2x covenant requirements; Particularly broadly or loosely worded MAC clauses exist, or one or more rating trigger exists within six notches of the current rating; or The insurer is at risk, within the next 12 months, of breaching one or more covenants because of a noneconomic deterioration in financial metrics primarily triggered by a change in accounting standards. 4. An insurer's liquidity ratio 217. The fourth and last liquidity subfactor assesses an insurer's ability, over a one-year period, to convert assets to cash, relative to the demand for its cash by policyholders and lenders. The proposed criteria use liquidity ratios to measure that ability The subfactor is "positive" when the ratio exceeds 2.2 times (x). For a life insurer or reinsurer, the subfactor is scored "neutral" when it is within x, and "negative" when it is less than 1.4x; for a P/C insurer or reinsurer, it is "neutral" when it is within x, and "negative" when it is less than 1.0x; for a multiline insurer, the cutoff points are blended according to the respective contributions of the life and P/C operations to stress insurance liabilities. The coverage is calculated based on our forward-looking view over the next 12 months, factoring in the insurer's liquidity management JULY 9,

50 and our cash flow expectations Given the differences in the characteristic of life versus P/C insurers and the different ways that insurance companies disclose data, the calculation of the liquidity ratio varies For U.S. and Canadian life insurance operating companies, the ratio is defined as the outcome of Standard & Poor's liquidity model (see "Criteria Insurance Life: Liquidity," published on April 22, 2004). The model measures the insurer's liquidity under two scenarios, immediate and ongoing stress, as the ratio of allowable assets divided by adjusted potential and maturing obligations. Under the proposed criteria, the insurer's liquidity ratio is the lower of the two ratios For all other insurers, the liquidity ratio addresses the extent to which the company could cover its stressed insurance liabilities outflows, defined in the following two paragraphs, with stressed liquid assets, which are liquid assets net of risk charges. These charges are 50% for listed equities, 10% for investment-grade bonds, 35% for bonds rated in the 'BB' and 'B' categories, 1% for deposits with banks rated above 'BBB-', and 5% for deposits with banks rated 'BB+' and below. Most other asset classes have a 100% charge including loans, private equity and hedge funds, property assets, and premium receivables. However, when material, certain entity-specific assets may be included provided that the insurers can demonstrate that it is possible to convert them promptly into cash. The applicable charge would be one of the above based on a review of its specific liquidity characteristics For P/C insurers, stressed claims outflows factor in stressed claims reserves on claims reserve duration. The liquidity ratio is defined as: Stressed liquid assets/[(net claims reserves + net reserve risk charge)/duration) + net catastrophe charge + net premium charge]. The claims reserves duration reflects the "tail" of the business underwritten. This assessment reflects the insurer's mean term of claim of liabilities, as referred to in the capital adequacy model criteria (see 52). Stress claims outflows include the impact of the property catastrophe risk charge and the P/C reserve and premium risk charges from the capital adequacy model For life insurers outside North America, stressed claims outflows factor in abnormally high lapse levels. The liquidity ratio is defined as stressed liquid assets divided by 35% of the sum of lapsable and transferable life liabilities. Linked business is excluded and assessed separately based on the liquidity of underlying assets. Based on global experience, the proposed criteria refer to an abnormally high lapse level as 35% of lapsable and transferable life liabilities. Examples of such liabilities include all continental Europe participating business, annuity liabilities, and with-profit liabilities. E3. Fixed-Charge Cover Test 224. An insurer needs to meet a minimum level of coverage for each SACP or GCP category (see table 26). For example, if the cover expectation is 4x, the SACP or GCP is capped at 'aa-'; 'aa' would require 5x or more. JULY 9,

51 Table 26 Fixed-Charge Coverage Test Of SACP And GCPs SACP or GCP Minimum EBITDA/fixed-charge coverage aaa 8 times (x) aa 5x a 3x bbb 2x bb 1.5x GCP--Group credit profile. SACP--Stand-alone credit profile. E4. Rating An Insurer Above The Sovereign Rating Or T&C Assessment 225. The proposed criteria may, in rare instances, result in an SACP or GCP--and potentially a rating--on a domestic unsupported insurer that is one to two notches above the local currency rating on the sovereign in whose jurisdiction the company has most of its business (see "Factoring Country Risk Into Insurer Financial Strength Ratings," published Feb. 11, 2003). In these cases, the proposed criteria typically subjects the SACP or GCP that results from sections A through E3 to the test in 228. In rating an insurer above the sovereign, Standard & Poor's is expressing its view that the company's willingness and ability to service debt is superior to the sovereign's and that, ultimately, if the sovereign defaults, there is a measurable probability that the insurance company will not default If an insurer derives less than 10% of both its assets and policyholder liabilities from a jurisdiction, including that of its domicile (for example, certain Ireland-, Bermuda- or Cayman Island-based insurers), its ratings are neither capped nor directly linked to the sovereign rating on that jurisdiction If the insurer is based in the European Economic and Monetary Union (EMU), the criteria in "Nonsovereign Ratings That Exceed EMU Sovereign Ratings: Methodology And Assumptions," published on June 14, 2011, continue to apply In all other cases, the following test applies. The local currency ratings on a domestic unsupported insurer could exceed the sovereign foreign currency rating level, by up to two notches only in rare circumstances, where insurer's specific strengths, beyond the general country risks factored in IICRA, sufficiently offset the following three types of risks: Asset risk, given not only the typically very high proportion of government securities on insurers' balance sheets, but also the impact of sovereign, and often macroeconomic, stress on real estate and corporate equity and debt values; Regulatory risk, because in our historical experience, including in Argentina in , regulatory frameworks and surveillance may well become significantly credit-negative as a sovereign undergoes credit stress and most insurers, including members of nondomestically owned groups, are subject to local regulations; and Potential direct government intervention, mandated changes in the contractual terms of debt or insurance obligations, or by other means, for example, in response to an economic crisis associated with sovereign credit distress In addition, the government local currency rating typically caps the ICR and FSR on a domestic unsupported insurer. The insurer's foreign currency rating (whether FSR or ICR) is the lower of its local currency rating and the relevant JULY 9,

52 T&C assessment. While FSRs only address policyholder obligations and T&C assessments refer to debt service, the proposed criteria equalize the risk that a sovereign would apply restrictions on the availability of foreign exchange to service debt with the risk that the sovereign would apply restrictions on the availability of foreign exchange to service policyholder obligations For the purpose of this section, "domestic" refers to an insurer where most (typically 70% or more) of its assets or policyholder obligations, whichever proportion is higher, stem from a given jurisdiction or its jurisdiction of domicile. "Unsupported" means that the insurer is not guaranteed by, nor "core" or "highly strategic" to, a nondomestic group The ratings on a domestic insurer that is guaranteed by, or a branch of, a rated nondomestic parent or affiliate, are the lower of the latter's ratings and the level six notches above the local currency rating, if investment grade (four if speculative grade), of the sovereign where the domestic insurer is domiciled. Generally, senior group management would have demonstrated a strong commitment, including a track record of support in good times as well as bad, for the subsidiary or branch For insurers that are not based in the EMU and do not meet the conditions in 226 or 230, the assessment looks at the blended exposure of T&C, sovereign, and country risk levels. F. Support Framework 233. The proposed criteria base the ICRs of members of insurance groups, both operating and holding companies, on the following: For an insurance operating company: its group status; its SACP, according to sections A to E above; and the GCP. For a nonoperating insurance holding company (NOHC): the GCP and certain factors that drive the differential between the GCP and the holding company's ICR. A NOHC is defined as a company that does not conduct material insurance operations itself but is the ultimate or an intermediate group owner whose unconsolidated assets predominantly (generally more than 90%) comprise investments in, or amounts due from, its subsidiaries. For an operating holding company: the GCP, notched down to the extent that its holding company activities outweigh its operating company activities. For a noninsurance operating company: its group status, its SACP assigned according to the relevant criteria for its type (bank, other financial institution, or corporate), and the GCP. If the insurer is a GRE, its SACP incorporates the ongoing aspects of the relationship with the related government; its GCP incorporates our expectations of potential extraordinary government support or negative intervention should the insurer come under stress The FSR on an insurance operating company is set at the same level as the ICR on the company except: To reflect explicit support for policyholder obligations (a guarantee or net-worth maintenance agreement, see 242 to 245). Since the support does not extend to debtholders, the FSR may be enhanced by such support, whereas the ICR is not. When ratings are in the 'B' category or below, where the ability and willingness to service debt may differ from the ability and willingness to service policyholder obligations, the gap between the two ratings may span several notches. JULY 9,

53 JULY 9,

54 F1. Rating Insurance Subsidiaries Of Insurance Groups 1. Section applicable to all members of insurance groups 235. The proposed criteria base the rating of an insurance group's subsidiary on the Methodology section (Section III) of "Group Rating Methodology And Assumptions," published Nov. 9, 2011 (the GRM criteria), and the insurance-specific criteria in this section. That methodology explains how we define a group, derive a GCP, assess the group status of each rated group member, and how the subsidiary's rating is derived from the combination of the GCP, its group status and, when the latter is "strategically important" or of lesser importance, its SACP (see chart 2, where the paragraph numbers refer to this RFC) Supplementing 28 of the GRM criteria, for "core" and "highly strategic" subsidiaries of insurance groups, other than captive insurers, "commensurate capitalization" refers to a capitalization that is in line with group policies and practices for key subsidiaries and stands significantly above the MCR Supplementing 27 of the GRM criteria, for "core," "highly strategic," and "strategically important" subsidiaries of insurance groups, other than captive insurers, a "significant proportion" refers to at least 5% of both capital and consolidated operating earnings. For the purpose of this analysis, "capital" is measured by TAC and "operating earnings" by "adjusted EBIT" (see "Assumptions For Quantitative Metrics Used In Rating Insurers Globally," April 14, 2011, 31). The metrics take into account current figures and projections for the next two years based on the recent track record Supplementing 30 of the GRM criteria, for insurance groups, a group subsidiary that is not a captive insurer is "highly strategic" if it meets all "core" characteristics except for "either constitutes a significant proportion of the consolidated group or are fully integrated with the group." 239. Supplementing 27 to 31 of the GRM criteria, an insurance group's subsidiary is not considered "core," "highly strategic," or "strategically important" if there is any possibility of it being placed into run-off. However, this does not apply to subsidiaries whose operations could be transferred to other core, highly strategic, or strategically important subsidiaries, as long as there is no measurable credit impact on policyholder and nonpolicyholder financial obligations. In addition, this does not apply to subsidiaries of groups that for reputation reasons will likely support a subsidiary even in run-off, or which continue to consider as strategic the line of business to which the subsidiary contributes Supplementing 27 of the GRM criteria, an insurance group's newly acquired subsidiary may not usually be designated as "core" during the first two years after the acquisition because of integration risks and the potential for new, unanticipated risks to emerge. A subsidiary may usually be designated as core after it is fully integrated. However, significant and sustained deterioration of operations or earnings underperformance could also cause a reclassification of its group status to a lower category than highly strategic Section B4, 55 to 61, of the GRM criteria regarding "insulated subsidiaries" would apply to an insurance group, with certain exceptions. In 60, the dividend policy is analyzed according to all publicly available information. JULY 9,

55 2. Insurance subsidiaries as beneficiaries of policy guarantees and other support agreements 242. Where a policy guarantee agreement meets the following conditions, the FSR on the beneficiary is that of the guarantor (unless the beneficiary's SACP is higher). These conditions mirror those for our rating substitution criteria (see "Guarantee Default: Assessing The Impact On The Guarantor's Issuer Credit Rating," May 11, 2012), except that the last two are specific to the proposed criteria, as is the absence of a reference to timeliness (which FSRs do not address). Also, policyholders, not debtholders, are the beneficiaries. The guarantee covers all policyholder obligations and explicitly ranks them as pari passu with the guarantor's own policyholder obligations. (A guarantee that does not cover all policyholder obligations of the guaranteed entity may not enhance the latter's FSR at all.) The guarantee is of payment and not collection. The guarantee is unconditional, irrespective of value, genuineness, validity, or enforceability of the supported obligations. The guarantee provides that the guarantor waives any other circumstance or condition that would normally release a support provider from its obligations. The guarantor also should waive the right of set-off and counterclaim. The guarantor's right to terminate the agreement is appropriately restricted, i.e., the support agreement does not terminate before the supported obligations are paid in full. Alternatively, if it does, obligations incurred up to the termination date remain supported. In addition, the support agreement must be binding on successors and assigns of the support provider or, if it can be revoked, this only applies to policies written after the revocation date. The guarantee provides that it reinstates if any supported payment is recaptured as a result of the primary obligor's or the guarantor's bankruptcy or insolvency. Policyholders are third-party beneficiaries of the guarantee. In the case of cross-border transactions, the risk of withholding tax with respect to payments by the guarantor may need to be addressed. In addition, the guarantor typically must subject itself to jurisdiction and service of process in the jurisdiction in which the guarantee is to be performed. To strengthen the guarantee's enforceability by policyholders, if it is not referenced in insurance policies, the beneficiary insurer provides sufficient public disclosure of its existence and key features For the purpose of these criteria, "support agreements" include net-worth maintenance agreements or any other agreement intended to provide support to subsidiary policyholders. Where a support agreement does not meet all of the conditions in the previous 242, to qualify for any rating enhancement, the support agreement must meet the following conditions in addition to those in 244 and 245: It gives policyholders, financial creditors, or other third-party interests, such as regulators, the ability to enforce the agreement against the support provider, if the provider fails to perform its obligations. It cannot be modified or terminated to the detriment of the existing beneficiary policyholders, or creditors at the time of termination without their agreement, unless the beneficiary subsidiary's creditworthiness becomes at least as strong as the supported rating; or the beneficiary can be sold only to an insurer with the same or higher creditworthiness as the support provider. It stipulates that the subsidiary will be prudently capitalized, for example, relative to the regulatory capital requirement. It provides that the support provider will cause the beneficiary entity to have sufficient cash and liquid assets for the timely payment of all of its debt if the agreement is to provide corporate debt support and policyholder obligations if the agreement is to provide policy support. JULY 9,

56 244. Where, in addition to the conditions in the previous paragraph, the beneficiary subsidiary is at least "strategically important" to the group and the support agreement meets all of the following four conditions, the FSR on the beneficiary (unless it has an SACP above the GCP) is the same as the rating on the support provider: The agreement states definitively that the provider will support the beneficiary, and sets no material cap on the support; The agreement is provided by a regulated bank or insurer that is a core group member; The agreement is binding on successors and assigns of the support provider; and The beneficiary subsidiary does not demonstrate adverse performance and is not likely to be part of a corporate restructuring. If the other four conditions above are met, but not this one, the rating on the beneficiary would be set one notch below the GCP, unless its SACP is the same as the GCP Where in addition to the conditions in 243, a net-worth maintenance agreement meets both of the following conditions, the rating on the beneficiary is raised by three notches from its SACP, subject to a cap at one notch below the support provider's rating. The agreement demonstrates a current intention to support the beneficiary in the medium to long term; and The agreement is provided by an affiliated regulated bank or insurer. F2. Assigning ICRs To Nonoperating Holding Companies 246. This section addresses how ICRs on NOHCs are derived from the GCP. Holding companies that meet the conditions for core group status are operating holding companies; other holding companies are NOHCs. NOHCs are not assigned FSRs, and GRM designations are not applied to these companies A NOHC's ICR would be determined by (1) the GCP and (2) the number of notches by which it would be differentiated to reflect ongoing cash flow subordination between the creditors of the holding company and those of the operating insurance subsidiaries (typically their policyholders) (see table 27). JULY 9,

57 248. The NOHC's ICR notching relative to the GCP reflects the degree of structural subordination within insurance groups. Structural subordination is considered very high in jurisdictions such as the U.S., where even strong companies have to obtain prior regulatory approval before transferring significant amounts of solvency capital from an operating company to its holding company. Structural subordination is somewhat less onerous in regions such as the EU The NOHC's liquidity assessment is a function of the first three of the four subfactors defined in section E2 and of the ratio subfactor described in 252, all analyzed at the level of the unconsolidated holding company, which, in most cases, bears most of the group's financial obligations Liquidity is scored "strong," "adequate," "less than adequate," or "weak." 251. Liquidity is scored "strong" when no subfactor is negative and at least two are positive; "less than adequate" when one or two are negative; and "weak" when three or four are. In all other cases, it is "adequate." 252. The ratio subfactor is positive when both ratios exceed 1.5x, negative if the first one is under 1.2x and the second one under 1.0x, and neutral otherwise. Liquid assets to noncontingent short-term financial liabilities, where the numerator excludes stakes in subsidiaries and liquidity facilities, and the denominator includes liabilities with structured settlements, with no optional features. The holding company's ability to pay its total liquidity requirements (excluding principal servicing) out of its cash inflows: [Dividends from operating entities + net investment revenues from holding assets] / [overhead expenses + interest charges + other ongoing financial charges + shareholder distributions, if any] The notching in table 27 is increased in the following situations: JULY 9,

58 When the holding company's liquidity assessment is "less than adequate" or "weak." Its ratings are then capped at 'BB+' or 'B-', respectively. When the holding company itself carries very significant asset or liability risks that are otherwise diluted within the overall GCP. F3. Assigning Issue Ratings 254. This section addresses how to assign ratings to long-term obligations that are not deferrable or mandatorily convertible and are issued or guaranteed by insurers that are members of insurance groups Obligations with a guarantee that meets our rating substitution criteria are rated at the guarantor's level (if severally guaranteed: at the highest of all guarantors' ratings) Obligations that do not benefit from such guarantees are rated according to table 28. Even if the ICR is supported by a parent, affiliate, or government, table 28 applies, since the ICR is weak-linked to each of these obligations in the first place. Table 28 Determining Issue Ratings On Nondeferrable Obligations from ICRs* Situation Obligation type Typical notching Holding company, ICR is investment grade Holding company, ICR is speculative grade Operating company, ICR is investment grade Operating company, ICR is speculative grade *Nondeferrable also encompasses mandatory-convertible securities. Senior debt 0 Junior debt -1 Senior debt 0 Junior debt -2 Senior debt -1 Junior debt -1 Senior debt -2 Junior debt -2 Junior obligation notching could be one notch less in rare cases where recovery prospects would be unusually strong, for example if we expect capitalization to remain stronger in a default scenario than our usual expectations. Senior debt could (in cases that are expected to be rare when the ICR is 'B-' or higher) be rated one notch higher than the ICR if holders benefit from asset securities that considerably enhance recovery. Except if a recovery rating is assigned, in which case the notching would reflect the recovery rating according to "Recovery Ratings On The Debt Of Speculative-Grade Companies In The Insurance Sector," June 24, Reflects typical policyholder seniority over financial lenders. Notching is zero (a) in the rare jurisdictions where policyholders would not be senior to financial lenders or (b) for very well-secured senior debt of speculative-grade companies. JULY 9,

59 VIII. GLOSSARY BRP. An insurer's business risk profile. Capital model and capital model criteria. The capital model is a quantitative tool that is integral to Standard & Poor's analysis of the capital adequacy of life, property/casualty, health insurance, and reinsurance companies worldwide, as described in the criteria article "Refined Methodology and Assumptions For Analyzing Insurer Capital Adequacy Using the Risk-Based Insurance Capital Model," published on June 7, 2010, as supplemented by "A New Level Of Enterprise Risk Management Analysis: Methodology For Assessing Insurers' Economic Capital Models," Jan. 24, to 30 in this article describe our use of U.S. generally accepted accounting principles (GAAP), statutory accounts, International Financial Reporting Standards (IFRS), and local GAAP on a consolidated, aggregated, or unconsolidated basis depending on whether we conduct the analysis on a group or subsidiary level. The capital model derives a risk-based capital (RBC) requirement amount at four confidence levels (see 21-22). Coinsurance. Insurance (or reinsurance) business where insurers share the same terms and conditions as other insurers underwriting the same risk, other than the proportion of that risk. For example, insurer A may insure 40% of the risk, and insurers B and C may each insure 30% of the risk. In this example, the insurers would normally share premiums, commissions, and claims in the same proportions. Counterparty credit rating (CCR). This is the same as issuer credit rating (ICR). FER or financial enhancement rating. A FER addresses an insurer's ability and willingness to meet credit-enhancement insurance claims on a full and timely basis. See "Financial Enhancement Ratings," published Dec. 10, 2004, and "Standard & Poor's Ratings Definitions," published June 22, FRP. An insurer's financial risk profile. FSR or financial strength rating. A Standard & Poor's insurer financial strength rating is a forward-looking opinion about the financial security characteristics of an insurance organization with respect to its ability to pay under its insurance policies and contracts in accordance with their terms. For organizations with cross-border or multinational operations, including those conducted by subsidiaries or branch offices, the ratings do not take into account potential that may exist for foreign exchange restrictions to prevent financial obligations from being met. See "Standard & Poor's Ratings Definitions," published June 22, However, under the proposed criteria we also would assign a foreign currency FSR when it is different than an insurer's local currency FSR. For a given insurer, the ICR and FSR might differ when one class of the insurer's obligations is guaranteed by an entity with a different ICR; or when the insurer is in distress or in default. Typically FSRs are not assigned to nonoperating holding companies because they don't underwrite insurance risks. GCP or group credit profile. The GCP is Standard & Poor's opinion of a group's creditworthiness as if the group were a single legal entity, and is conceptually equivalent to an ICR. A GCP does not address any specific obligation. See "Group Rating Methodology And Assumptions," Nov. 9, 2011, 11. GRM or group rating methodology. Methodology to assess financial support within a group (see "Group Rating Methodology And Assumptions," Nov. 9, 2011). Hard market. A period when premiums that insurers charge are high relative to long-term norms. Hybrid instruments. These securities, which include preferred shares, combine features of debt and equity, but are not equivalent to common equity or senior debt. ICR or issuer credit rating. A Standard & Poor's issuer credit rating is a forward-looking opinion about an obligor's overall creditworthiness in order to pay its financial obligations. This opinion focuses on the obligor's capacity and willingness to meet its financial commitments as they come due. It does not apply to any specific financial obligation, as it does not take into account the nature of and provisions of the obligation, its standing in bankruptcy or liquidation, statutory preferences, or the legality and enforceability of the obligation. See: "Standard & Poor's JULY 9,

60 Ratings Definitions," published June 22, Also called "counterparty credit rating." An insurer is assigned both foreign and local currency ICRs. Insurance or Insurers. In these criteria, unless otherwise stated, these terms include reinsurance and reinsurers. Insurance group. A group of companies that have insurance as their predominant activity. Local currency issuer credit rating. A nonsovereign entity's local currency ICR reflects Standard & Poor's opinion of that entity's willingness and ability to service its financial obligations, regardless of currency and in the absence of restrictions on its access to foreign exchange needed to service debt. Minority interests. Also referred to as noncontrolling interests. New business margin. The ratio of value of new business divided by (1) the present value of new premiums or (2) if the present value of new premiums is not available, by the sum of (i) new annual premiums and (ii) 10% of single premiums. The value of new business is the present value of the future profits of all new policies sold during the year. P/C or property/casualty. Premiums. For insurers reporting under U.S. generally accepted accounting principles, given that under Financial Accounting Standards Board Statement No. 97 most annuities and universal life receipts are accounted for as deposits and not as revenues or premiums, for the purpose of these criteria premiums are represented by sales. RBC or risk-based capital. ROTC or return on total capital. We define ROTC as (a) the sum of (i) net income and (ii) interest expense multiplied by (1 effective tax rate), divided by (b) the sum of reported equity, hybrid, and debt. The numerator is the average over the period and the denominator is average between the beginning and the end of the period. Sigma. This refers to the study, Sigma Study: No. 3/2012, "World insurance in 2011: non-life ready for take-off." Soft markets. These are periods when premiums charged are believed to be low relative to long-term norms. Solvency margin. This is the amount by which an insurance company's assets exceeds its projected liabilities, effectively a measure of its financial health. TAC or total adjusted capital. TAC is the measure Standard & Poor's uses to define the capital available to meet a company's capital requirements in our capital adequacy model, as derived from our capital adequacy model ("Refined Methodology and Assumptions For Analyzing Insurer Capital Adequacy Using the Risk-Based Insurance Capital Model," June 7, 2010, table 1.) T&C: Transfer and convertibility, as defined in "Criteria For Determining Transfer And Convertibility Assessments," published May 18, A T&C assessment is the rating associated with the likelihood of the sovereign restricting access to foreign exchange needed for debt service. Tied (otherwise known as exclusive) agents and non-tied agents. Tied agents are those that are contractually bound to distribute the products of only one insurer. They may not be controlled by the insurer, but they are significantly influenced by the insurer because of the exclusivity of the relationship. For the purpose of these proposed criteria, the agent may be tied to different insurers for different products but these products must not be substitutes for each other. For example, if the agent's customer requires home insurance or term life insurance, it may only offer the product of one insurer in each case. Non-tied agents are all other agents. IX. RELATED CRITERIA AND RESEARCH Related but would not be superseded, even in part Principles Of Credit Ratings, Feb. 16, 2011 Standard & Poor's Ratings Definitions, June 22, 2012 Credit Stability Criteria, May 3, 2010 Understanding Standard & Poor's Rating Definitions, June 3, JULY 9,

61 Criteria For Determining Transfer And Convertibility Assessments, May 18, 2009 Nonsovereign Ratings That Exceed EMU Sovereign Ratings: Methodology And Assumptions, June 14, Rating Government-Related Entities: Methodology And Assumptions, Dec. 9, 2010 Stand-Alone Credit Profiles: One Component Of A Rating, Oct. 1, 2010 How Standard & Poor's Uses Its 'CCC' Rating, Dec. 12, Refined Methodology And Assumptions For Analyzing Insurer Capital Adequacy Using The Risk-Based Insurance Capital Model, June 7, Life: Liquidity, April 22, A New Level Of Enterprise Risk Management Analysis: Methodology For Assessing Insurers' Economic Capital Models, Jan. 24, Assumptions for Quantitative Metrics Used in Rating Insurers Globally, April 14, Would be partly superseded Group Rating Methodology And Assumptions, Nov. 9, Holding Company Analysis, June 11, Factoring Country Risk Into Insurer Financial Strength Ratings, Feb. 11, Group Methodology, April 22, JULY 9,

62 Copyright 2012 by Standard & Poor's Financial Services LLC. All rights reserved. No content (including ratings, credit-related analyses and data, model, software or other application or output therefrom) or any part thereof (Content) may be modified, reverse engineered, reproduced or distributed in any form by any means, or stored in a database or retrieval system, without the prior written permission of Standard & Poor's Financial Services LLC or its affiliates (collectively, S&P). The Content shall not be used for any unlawful or unauthorized purposes. S&P and any third-party providers, as well as their directors, officers, shareholders, employees or agents (collectively S&P Parties) do not guarantee the accuracy, completeness, timeliness or availability of the Content. S&P Parties are not responsible for any errors or omissions (negligent or otherwise), regardless of the cause, for the results obtained from the use of the Content, or for the security or maintenance of any data input by the user. The Content is provided on an "as is" basis. S&P PARTIES DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OR USE, FREEDOM FROM BUGS, SOFTWARE ERRORS OR DEFECTS, THAT THE CONTENT'S FUNCTIONING WILL BE UNINTERRUPTED, OR THAT THE CONTENT WILL OPERATE WITH ANY SOFTWARE OR HARDWARE CONFIGURATION. In no event shall S&P Parties be liable to any party for any direct, indirect, incidental, exemplary, compensatory, punitive, special or consequential damages, costs, expenses, legal fees, or losses (including, without limitation, lost income or lost profits and opportunity costs or losses caused by negligence) in connection with any use of the Content even if advised of the possibility of such damages. Credit-related and other analyses, including ratings, and statements in the Content are statements of opinion as of the date they are expressed and not statements of fact. S&P's opinions, analyses, and rating acknowledgment decisions (described below) are not recommendations to purchase, hold, or sell any securities or to make any investment decisions, and do not address the suitability of any security. S&P assumes no obligation to update the Content following publication in any form or format. The Content should not be relied on and is not a substitute for the skill, judgment and experience of the user, its management, employees, advisors and/or clients when making investment and other business decisions. S&P does not act as a fiduciary or an investment advisor except where registered as such. While S&P has obtained information from sources it believes to be reliable, S&P does not perform an audit and undertakes no duty of due diligence or independent verification of any information it receives. To the extent that regulatory authorities allow a rating agency to acknowledge in one jurisdiction a rating issued in another jurisdiction for certain regulatory purposes, S&P reserves the right to assign, withdraw, or suspend such acknowledgement at any time and in its sole discretion. S&P Parties disclaim any duty whatsoever arising out of the assignment, withdrawal, or suspension of an acknowledgment as well as any liability for any damage alleged to have been suffered on account thereof. S&P keeps certain activities of its business units separate from each other in order to preserve the independence and objectivity of their respective activities. As a result, certain business units of S&P may have information that is not available to other S&P business units. S&P has established policies and procedures to maintain the confidentiality of certain nonpublic information received in connection with each analytical process. S&P may receive compensation for its ratings and certain analyses, normally from issuers or underwriters of securities or from obligors. S&P reserves the right to disseminate its opinions and analyses. S&P's public ratings and analyses are made available on its Web sites, (free of charge), and and (subscription), and may be distributed through other means, including via S&P publications and third-party redistributors. Additional information about our ratings fees is available at JULY 9,