National Harbor, Md.

State insurance regulators, after a hearing examining the flawedperformance of top rating agencies in assessing the risk of complexsecurities held by carriers, left unresolved what their next movewill be, holding open the possibility of performing the taskthemselves.

A trio of rating agencies–Moody's, Fitch and Standard &Poor's–were called on the carpet during a day-long hearing on Sept.17 at the Fall National Meeting of the National Association ofInsurance Commissioners, which employs rating agency evaluations todetermine insurer risk-based capital requirements.

During theproceedings, regulators discussed whether rating agency failures todetect the risks of mortgage-backed securities and otherderivatives were caused by conflicts of interest related to theagencies' for-profit business model, or a misunderstanding of theexposures associated with the structured securities theyassessed.

Commissioners at the NAIC's Rating Agency Insurance WorkingGroup agreed their level of reliance on rating agencies needs to bere-examined, but left unclear what alternatives might emerge.

Plans for the NAIC to take on rating responsibilities in-houseremain on the table, although Chris Evangel, managing director ofthe NAIC Securities Valuation Office, cited in his testimony somelogistical difficulties with that strategy, as well as costconcerns.

Still, commissioners said that for now, every option will beconsidered. "We owe it to consumers to be thoughtful anddeliberate," Connecticut Insurance Commissioner Tom Sullivansaid.

As for where rating agencies went wrong in the credit crisis,those testifying at the hearing disagreed.

The rating agencies themselves admitted they erred in ratingresidential mortgage-backed securities as sound, but said that wasthe exception in a long history of success.

Agency representatives said they have taken steps in the areasof governance, compliance and an institutional rotation of analyststo guard against future failures.

However, they also said the information they provide has beenmisused, or even used with a degree of na?vet?, by those who hadcome to rely on their services.

A former employee of Moody's–Jerome Fons, principal with FonsRisk Solutions, a New York-based credit risk consultingfirm–suggested in his testimony that the rating agencies'for-profit model creates a conflict of interest that contributed tothe current crisis. He said rating agencies "traded theirreputation for profits."

However, David Marks, executive vice president and chiefinvestment officer of Madison, Wis.-based CUNA Mutual Group, saidhe does not believe it was the for-profit model, but rather a lackof understanding of the types of structured securities being ratedthat created the rating agencies' missteps.

Commissioners were divided as well.

Mr. Sullivan said he sees inherent conflicts in how ratingagencies operate, adding that the for-profit model, combined withthe pressure within rating agencies to be the first to upgrade andfirst to downgrade, creates a "conflicted model" that has evolvedover time and cannot be changed overnight.

"While I hear what rating agencies are saying about governanceand compliance and all those other nice things, until you changethe structural underpinnings of how they operate, I don't think youcan fundamentally change how they rate," he said.

Mr. Sullivan also challenged the notion that regulators havemisused the ratings, accusing the rating agencies of "doubletalk."

"On the one hand, they're saying you shouldn't rely so heavilyon us, and then on the other hand they're saying, 'We've beenaround for decades and our reputation is everything to us.' That tome is a little bit of double-talk," said Mr. Sullivan

New York Insurance Superintendent James Wrynn, co-chair of theRating Agency Working Group, attributed rating agency failures moreto a misunderstanding of what it was they were rating.

He said regulators will look at the issuer-pay model of therating agencies, "but I wouldn't say that was the primary cause ofthe problem. I think it's more the nature and combination offactors that converged at once."

Mr. Wrynn noted the "inability to see what was going on with theunderlying assets of these structured securities."

Regarding future use of rating agencies, he said he still seesthem as a tool–just not the only tool. "There really wasn't anissue with the rating agencies until 2000," when corporate failuressuch as Worldcom arose, according to Mr. Wrynn, referring to thetelecommunications giant that went bankrupt after an accountingscandal.

Mr. Wrynn said rating agencies will have to work to restore thepublic's confidence.

Illinois Insurance Director Michael McRaith, co-chair of theRating Agency Working Group, said he was unsure whether the problemrested with the rating agencies' for-profit model ormisunderstanding the structured securities they were rating. But hedid say rating agency representatives were not forthcoming withtheir responses.

"It's fair to say that as regulators we were frustrated by theexplanation from the rating agencies in response to that question,"Mr. McRaith said.

"I think it's fair to assume, because it wasn't said, that itwas a failure of methodology, it was a failure of leadership andgovernance, and ultimately it was a failure to acknowledge what washappening in the world," he added, noting that economists andothers raised red flags about residential mortgage-backedsecurities well before rating agencies took any action.

"There was not a lot of fact-based comment about what happenedand how we got here," he said. "It was disappointing not to hearthe rating agencies offer more reflection on their failures."

As for their next steps on how regulators should use ratings tohelp ensure the solvency of insurers, testimony was given on arange of options that included establishing an independent,nonprofit rating entity that would be funded by insurancecompanies, and promoting more competition among rating agencies bybreaking the virtual monopoly of the "big three"–Standard &Poor's, Moody's and Fitch.

Mr. McRaith asked multiple witnesses how much weight regulatorsshould give to ratings from the big three.

Matt Richardson, the Charles E. Simon professor of financialeconomics and director of the Salomon Center at New YorkUniversity, said rating agency figures have value as long as theirmodel is fixed first.

He said the models are imprecise now because of the complexityof the products they are modeling. These products are moresensitive to the initial values used in the models, he said, anddifferent input at the outset will cause wide variety in theresults.

Several who testified noted the difficulty in effectively ratingnewer structured securities that emerged in the marketplace asopposed to corporate debt.

Mr. Richardson also said rating agencies do not model some ofthe information regulators are looking for. Revised models canmeasure default risk, he said, but as of now they cannot modelliquidity risk and market risk.

Mr. McRaith said interested parties have until Oct. 7 to submitcomments before regulators make a decision on how to moveforward.