S&P - Public Finance Criteria (2007). - The Global Clearinghouse

Bond Insurancemargin of safety measurement provides greater insight.The margin of safety expresses ending capital in thecontext of the scale of the company. The margin ofsafety accomplishes this by relating total claims-payingresources (ending statutory capital plus losses) to losses.Thus, a margin of safety of 1.25x signifies thatending capital exceeded losses by 25%. Stated anotherway, losses could have been 25% larger without drivingthe statutory capital below zero.The margin of safety is a useful tool that allowsfor analysis of capital adequacy trends for individualcompanies and capital strength comparisons amongthe different insurers. Some bond insurers use themeasurement for capital planning purposes. Theminimum margin of safety for ‘AAA’ rated bondinsurers is 1.25x. For ‘AA’ and ‘A’ rated insurers theminimums are 1.0x and 0.8x, respectively. Theseminimum values can be adjusted slightly lower incases where the insurer is owned by a single highlyrated entity that has expressed continued support forthe company.Single-Risk Guidelines And AnalysisWhereas the capital adequacy model addressesthe question of capital relative to a severe,widescale claims-paying environment, single-riskstandards and analysis look at capital and ratingstability in the more likely context of occasionallarge discrete defaults by individual obligors. Aninordinately large exposure to a defaulting issueror issue could threaten a bond insurer’s rating,particularly in a nondepression environmentwhere the default is an isolated event and is notrelated to a general economic downturn.For this reason, Standard & Poor’s has insurerspecific,single-risk guidelines that limit exposuresto individual issuers or issues in the case of assetbackedtransactions. The approach is based on theassumption that any issuer or issue could suffer alarge discrete loss, despite investment-grade underwritingstandards, and measures the possible lossagainst the earnings power of the company.Investment-grade credits are not immune to default,and the single-risk standards reflect the furtherassumption that the severity of the loss will begreat, in the context of the obligor’s sector.The criteria for maximum single-risk exposure isbased on two key assumptions: (1) that the maximumloss allowable is a function of how much abond insurer could write off and still maintain itsexisting rating, and (2) that the expected loss onany issuer is a function of the issuer’s market sector.The loss tolerance (how much an insurer could loseand retain its rating) relating to a single issuer isequal to twice the company’s core earnings (seetable 5). Core earnings include adjustments fortaxes, advanced refundings, capital gains and losses,and nonrecurring income statement items.This approach conservatively identifies potentialearnings net of any nonrecurring items. Because anylarge loss would shelter a significant amount ofearnings from taxes, pretax earnings are used in thecalculation. In addition, since refunded earned premiumscan vary greatly, refunded earned premiumsfor the base year are compared with the lowest levelof premiums earned from refundings over the priorfive years. The lower amount is included in the coresingle-risk earnings calculation. This methodologynormalizes some of the income statement components(thereby reducing loss-tolerance variability)and facilitates the single-risk planning process.For unseasoned financial guarantors—those thathave yet to develop a significant level of core earnings—themaximum allowable exposure to a lossfrom a single issuer is expressed as a percent of originalsurplus. The percent used is equal to twice thepredictable, yet conservative, rate a seasoned bondinsurer could earn on its existing surplus for one year.Currently, a 12.5% rate of return is assumed for thesepurposes. Single-risk limits for unseasoned companiesremain based on original surplus adjusted for subsequentcapital infusions until core earnings are sufficientto generate a higher computed loss tolerance.The single-risk categories for each sector areshown in tables 6 and 7. Based on the relativedegree of risk between the categories and the earningspower of a seasoned company or the assumed12.5% rate of return for an unseasoned company,the maximum exposures to a single-risk by categoryare shown in table 5. These relationships imply thatCategory 3 obligations are considered to have twicethe loss potential of Category 1, while Category 6obligations are considered to have four times theloss potential of Category 1. In other words, thelower the risk sector, the greater the insured principalamount of debt that an insurer can cover relativeto its earnings or capital base.Single-risk loss potentials for ABS are determinedon a case-by-case basis using the samecredit-gap concept employed to determine capitalcharges. A company’s earnings power or capitalbase is used to determine its loss tolerance foreach transaction. ■www.standardandpoors.com311