WASHINGTON — Fannie Mae and Freddie Mac's experiments with selling credit risk to investors are a critical element of new plan to merge the two entities and move them out of conservatorship, but what form so-called credit risk transfers take could make a big difference.
To date, the Federal Housing Finance Agency and the government-sponsored enterprises have mostly focused on back-end risk transfers involving pools of existing loans. But a team of five academics and housing finance experts are arguing that front-end risk transfers, when loans are being priced and pooled, can form the basis of a new system.
"It takes advantage of the risk-sharing transactions that the Federal Housing Finance Agency is already pushing," said Jim Parrott a senior fellow at the Urban Institute and former White House economic advisor, one of the co-authors of the plan. "We can learn from these credit risk transfers and what mix of structures we want in a permanent system."
The success of front-end risk sharing is critical for the plan, which was released last month and crafted by Parrot, mortgage securitization pioneer Lewis Ranieri; Gene Sperling, a former director of the National Economic Council; economist Mark Zandi; and Barry Zigas of the Consumer Federation of America.
So far, policymakers appear skeptical of the idea.
But the authors of "A More Promising Road to GSE Reform" — which would merge Fannie and Freddie into a single government corporation called the National Mortgage Reinsurance Corp. — expect it might take several years to better understand risk sharing transactions and develop a market with the capacity to invest in them.
"The GSEs and FHFA have some practical and policy reservations about front-end deals," Parrott said. "But our hope is that they will experiment with then over time, so we can compare their performance to the backend structures."
One reason for the skepticism is that backend deals are easier to conduct than front-end deals. Only large or sophisticated institutions can participate in backend deals, according to David Stevens, president and chief executive of the Mortgage Bankers Association.
"It leaves everybody else on the sidelines," he said, and the consumer does not see any price benefit.
But front-end deals solve a problem that back-end credit risk transfers don't — namely how to price the credit risk clearly and appropriately.
"To the extent to which the FHFA and the GSEs move into front-end credit risk transfers will determine how much more transparent pricing is and what true credit risk costs," said Michael Stegman, a counselor to Treasury Secretary Jack Lew on housing issues, at a Financial Services Roundtable event on Wednesday.
Under the plan, in a front-end deal, a private mortgage insurance company or investor could bid on loans being pooled and take a first loss position of up to 50% of the loan-to-value ratio. That would leave the GSE with 50% of the remaining credit risk. It would take a catastrophic economic event that resulted in extremely high default rates before the GSE would be impacted financially.
The five authors of the GSE reform plan want the combined GSE to transfer all non-catastrophic credit risk to a broad range of private entities.
If policymakers want the GSEs to shed their risk and become catastrophic risk providers, "you need to take the risk down to 50% LTV or somewhere in that range," Stevens said. "The mortgage insurers can certainly do that in terms of upfront risk taking."
U.S. Mortgage Insurers also want the FHFA to conduct a pilot program involving front-end credit-risk transactions.
"With the backend deals, there has been very little transparency in terms of pricing and whether there may be some benefit to the borrower or the lender that might be achieved," said Lindsey Johnson, president and executive director of the insurance group. "On the front-end deals, you have bids on the credit risk. It would be competitive and transparent."
With deeper credit enhancement on loans, Fannie and Freddie could also reduce the guarantee fees and loan level price adjustments they normally charge because they sold the first loss position to investors. All mortgage lenders could benefit from front-end credit-risk transfers, according to MBA.
FHFA is planning to "conduct an analysis and assessment of front-end risk transfers this year," according to its 2016 scorecard. After conducting the necessary analysis and assessment, the GSEs will "take the appropriate steps to continue front-end transactions."
FHFA declined to comment for this story.
But some observers doubt the success of front-end risk transfers.
"I don't think it is realistic" to expect a merged GSE to sell off most of the first loss credit risk and retain the catastrophic risk, according to Scott Olson, executive director of Community Home Lenders Association. "I am not sure it is workable, particularly when the market gets turbulent."
Others see the plan as giving more power to large institutions.
The front-end risk sharing transactions are a "great deal for the big banks" as well the mortgage insurers and large investors, according to Joshua Rosner, managing director for Graham Fisher & Co.
Rosner said it is difficult to determine the results of the risk-sharing deals "unless we know the GSEs' real cost of equity capital." However, this benchmark is not available because Fannie and Freddie are in conservatorship and don't have any capital.
"At the end of the day that should be a benchmark for looking at the risk sharing deals: is it more expensive or less expensive than GSEs' cost of equity capital?" he said in the interview.
Both Olson and Rosner were doubtful that lawmakers, particularly in the GOP, would consider the plan advanced by the five academics.
"The proposal would require new and comprehensive GSE reform legislation that is highly unlikely to pass," Rosner said in an April 3 report.
But others said at the very least it is rekindling discussion.
"This paper is clearly getting a lot of debate and it's resparked a dialogue" said Stevens.
