Freddie Mac is preparing to break new ground in its credit risk-sharing deals by offering investors something that they’ve been unable to get their hands on until now — exposure to actual losses on mortgages.
Earlier this year, Freddie began offering investors the opportunity to buy the first-loss position as part of its Structured Agency Credit Risk series. Now, Freddie is giving investors the opportunity to secure actual loss positions for the first time.
“The actual loss STACR offering will be similar to recent STACR deals, however, instead of allocating losses to the debt notes based upon a fixed severity approach, losses will be allocated in this transaction based on the actual losses realized on the related reference obligations,” Freddie said in a release. “Freddie Mac will continue to sell the first loss and mezzanine tranches while retaining a vertical slice of each tranche sold.”
Freddie announced Monday that it intends to market its newest offering, STACR 2015-DNA1, on April 13.
"We think the market of the future, where increasing amounts of credit risk will be transferred to private investors, will be actual loss based and we are excited to begin that transition with the next STACR offering," said Mike Reynolds, Freddie Mac vice president of credit risk transfer.
Freddie noted that it began making loan-level loss data available to investors in November. At the time, Freddie said that it was making the data available in an effort to increase investor transparency, and expected the loan-level data to help investors build more accurate credit performance models in support of Freddie’s credit risk-sharing offerings.
According to Freddie, loan-level actual loss data was added to its single-family loan-level historical dataset, which covers approximately 17 million 30-year, fixed-rate, single-family mortgages originated between January 1, 1999, and June 30, 2013.
Credit Suisse (CS) will act as the structuring lead manager for the new STACR offering, with Citigroup (C) as a co-lead manager and joint bookrunner.
