First, the good news: mortgage bankers that process loans for pooling into GNMA securities will continue receiving a 44 basis point fee for servicing those mortgages.
As for the processing of Fannie Mae and Freddie Mac loans, right now the future is wide open. (That would be the bad news.)
The Federal Housing Finance Agency is continuing to gather comments on its proposal which dictates a strict “fee for service” (set dollar amount) on Fannie/Freddie loans, but also considers a 20 basis point minimum with a 5 basis point reserve for problem loans.
The FHFA has been toying with the idea of changing the traditional 25 bp minimum for well over a year now and finally issued a second working paper/proposal in September—one that offered just two choices: fee for service and the 20/5 bps option, an idea being promulgated by the Mortgage Bankers Association and some of its allies.
No one knows for certain which compensation model the agency will ultimately pick, but there are growing fears that FFS is now winning the race at the FHFA.
Not surprisingly, FHFA officials have declined to discuss publicly which way they are leaning, and up until a few months ago the general thinking in the industry—certainly among servicing brokers and advisors—was that after contemplating change, the agency planned to do nothing regarding the minimum. But times change or as one consultant noted: “Everyone thought this whole thing was dead. Not any more.”
Servicing analysts and advisors are generally happy to discuss the issue with the trade media as long as they are not identified. And it doesn't take a genius to realize that these very same brokers/analysts stand to lose a way of life if FFS becomes the “reg of the land.” Then again, if FFS does win out, it might result in more lenders pushing FHA/VA products because of the 44 bp servicing strip. Meanwhile, a new comment letter from United Capital Markets of Denver speaks volumes on behalf of servicing advisors everywhere. UCM CEO Austin Tilghman provided a copy of his letter to NMN but declined to discuss it.
Among other things, UCM lays out a whole laundry list of reasons why FFS is a bad idea, chief among them: It reduces “skin in the game” for servicers, it might have the effect of increasing prepayment speeds, and it would actually increase the counterparty risk for Fannie and Freddie because the MSRs (which the GSEs technically own) would lose value as collateral.
The prepayment speeds argument is interesting and hasn't been aired all that much. UCM believes that a FFS model would actually incentivize lenders to churn loans because the business of servicing (if FHFA has its way) would become barely profitable.
For the record, UCM likes the MBA 20/5 bp proposal because it maintains the “skin in the game” model and would be beneficial to both MBS investors and guarantors.
If the agency ultimately chooses FFS, UCM sees a world where mortgage processing jobs will be driven overseas at an accelerated pace as profit margins tighten even more. “Companies like Ocwen and Zenta have already established operations in India to perform more labor-intensive functions as lower costs,” Tilghman writes in his nine-page comment letter.
The FHFA proposal talks about higher FFS payments for troubled loans but offers no details or estimates whatsoever. UCM believes the current FFS that the agency is pushing for performing loans doesn't allow mortgage bankers to recoup their basic costs.
Many servicers have complained that the way things stand now it's possible that mortgage bankers might even be incentivized to let loans go delinquent because that's when the higher fees kick in. If true, that doesn't sound like a particularly bright idea.
