The closely watched pot on my stove for the last six months, otherwise known as the Fed's MBS purchase program, is coming to an imminent end. (If you start here, you should be able to follow my posts on the subject as far back as you like. ) If you are up-to-date on the issues surrounding the Fed's departure (or insist, like an old school journalist, on getting the gist of the whole story in the lead), jump ahead to "Buyouts a Strong Technical." Otherwise, here's the brief:
  • Fed purchases since January 2009 consumed most of the new pass-through supply coming into the market from Fannie and Freddie (and a chunk of Ginnie's too);
  • Its demand has been a powerful tractor-beam pulling the spread between pass-through yields and mortgage rates over other high quality debt instruments to historic lows;
  • Removing that demand could allow pass-through yields and mortgage rates to widen dramatically;
  • MBS analysts had been estimating that traditional buyers would increase their demand, putting a "floor" on the widening, to around 20 to 40 basis points higher option-adjusted spreads. (OAS is used by the professionals who invest in MBS - you should not try this at home - because it extracts the expected effect of prepayments on realized yields over a range of interest rate scenarios.)
The future direction of mortgage rates, then, depends on the willingness of traditional MBS investors, including foreign investors, to return to the sector in the strength they displayed before 2008. (This is the subject of my March HousingWire Magazine column, "Who Takes the Slack From the Fed?") The one traditional sector not expected to take down its former "share" of pass-throughs is Fannie/Freddie. They have more pressing business to attend to - to buy out a delinquent loan pipeline they allowed to build rather than, under the old accounting regime, take the hit to capital (one the taxpayers would have had to make up). But avoiding that hit was a drain on earnings - one estimated at $15bn a year by J.P. Morgan Securities analysts. There is a limit on how much they can add to their portfolios as well. It was set by Congress in the Housing and Economic Recovery Act (HERA) of 2008 and declines 10% per annum beginning with this year. So, although they can - and Freddie has said it will - step in to sustain orderly markets for their securities, they cannot, as they would in the past, assume the role of the buyer on the margin. This was the mildly grim scenario analysts anticipated until last week. The GSEs had finally announced their buyout intentions and analysts could chew through to the consequences for securities and investors. Their first thoughts, on the day of the GSE announcements, February 10, were all about prepayments. But on second thought, by the end of last week, several were suggesting that the slug of prepaid cash will push traditional investors back into the MBS market. Buyouts a Strong Positive Technical Rates strategists Greg Reiter and Jeanna Curro at the Royal Bank of Scotland(RBS) in a report dated February 19, called the buyouts "a strong positive technical for the Agency MBS basis over the rest of 2010." (The basis is market slang for the spread between current coupon MBS and comparable high quality investments and benchmarks, typically Treasuries, interest rate swaps, and agency debt. In other words, a new tractor beam.) They estimate $166bn in cash will go back to investors who will reinvest "much" of it in agency MBS. But it's better than that. At the same time reinvestment demand is generated, MBS supply will be reduced by $166bn. Them's powerful technical forces. Furthermore, they point out, agency MBS may look rich on a short term basis, but against a longer historical perspective, they are "only a little rich." Net, net, the RBS team expects current coupon MBS spreads to widen by only 10 to 20 basis points. (By the way, congratulations are in order to RBS for building an MBS team under securitization veteran Brian Lancaster and kudos to Reiter and Curro, both highly respected for their work last decade at UBS, for getting the ball rolling so quickly. And good for us, as it's another positive sign essential capital markets services are being restored.) Laurie Goodman at Amherst Securities sees a substantial amount of that cash boosting CMO issuance. To explain - CMOs are a re-securitization, under the IRS' REMIC rules, of pass-throughs (or, back in the day, of pools of privately originated prime loans), that redistributes the interest rate and prepayment risk among different investor sectors having different risk preferences. For example, banks tend to own shorter bonds carved out of paydowns expected in the early years of the underlying pass-through terms, insurance companies bonds carved off intermediate- and last cash flows. Goodman points out that the short tranches of sequentially tranched CMOs,will receive all the buyout cash on the underlying pass-throughs. Banks, who own a disproportionately large share of these types of MBS, will have a disproportionate amount of cash to invest as a result. Goodman expects them to reinvest in new CMOs - driving large production of new issue CMOs over the next few months, focused in deals backed by the lower end of the coupon stack. Some of this demand has already been anticipated, she suggests, by the tightening in the 4.5 and 5.0% coupons. If a conventional (not Ginnie) coupon over 5% were to be used, Goodman expects it would be Freddies "which are post-buyout," though very well seasoned Fannies could be used as well. "In particular, most Fannies will be very difficult to use for CMO deals until the initial buyout program is largely complete. This should benefit the undervalued Gold/Fannie swaps." (Freddie's securities are nicknamed Golds. The point of the swap would be to buy Freddie, sell Fannie, and, when Freddie prices improve versus Fannie, reverse the trade. Or, expected Fannies to outperform, sell the swap. It is a fundamental trade of professional investors seeking to profit from short run expectations of market movements.) NOTE: Linda Lowell writes a regular column, called Kitchen Sink, for HousingWire magazine. Editor’s note: Linda Lowell is a 20-year-plus veteran of MBS and ABS research at a handful of Wall Street firms. She is currently principal of OffStreet Research LLC.