Pass-through investors are sitting on the edge of their chairs wondering how aggressively the GSEs are going to buy delinquent loans out of portfolio. Along with the potential for spread widening when the Fed retires its MBS purchase program, this is the main relative value consideration for agency MBS investors. Before we look for signs of buyout activity, let me bring everyone up-to-date on the questions. First, fundamentals: a critical determinant of pass-through investment performance is prepayments (which are slightly less predictable than interest rates!), and credit events like foreclosure, modification and buyout have become, in this environment, significant drivers of prepayments. Until loan modification efforts moved to center stage at the GSEs, investor apprehensions about credit prepayments were largely focused on GNMA pass-throughs, where servicer loan buyouts are the source of significant prepayment volatility (read: prepayment surges). GNMA servicers buy seriously delinquent loans out of pools across the coupon spectrum, but the practice is destructive when it occurs in pools carried at a premium to par. I discussed this practice in an earlier post. Last year, as modification efforts picked up across the GSE pass-through universe and HAMP came on line, buyout worries shifted to Fannie and Freddie pass-throughs. I walked through the issues after Christmas.
  • GSEs buyout rules allow them to buy a loan if it is delinquent 90 days or more, but they are obligated to do so if the loan is modified, a foreclosure sale occurs, or it is delinquent 24 months. They will also buy the loan out of a pool after 90 days if the cost of advancing P&I to investors exceeds the cost of holding the loan in portfolio.
  • Implementing FAS 167 changes the economic incentives. Before, a buyout hit capital hard, with maybe 40 to 60 points of market value loss, but starting Jan 1 2010 loans already on balance sheet. In effect, allowances for loan losses result in a "mark-to-model," not a market value hickey. Capital costs may be significantly moderated, especially at the time of buyout.
In the meantime, the cost of advancing on delinquent loans is high - J.P. Morgan analysts estimate it costs Fannie and Freddie about $15 billion a year to advance P&I on loans currently seriously delinquent. Given the cost and the change in accounting treatment, many MBS market analysts assumed the GSE's would move in the new year to buy loans out of pools as aggressively as possible subject to two hurdles: how to finance the purchases and portfolio limits set by Congress and expected to force the GSE's to shed 10% of year-end 2009 holdings by year-end 2010. Both questions were largely resolved - in my opinion and that of working MBS analysts - when the Treasury announced Christmas Eve that they were lifting the caps on capital support for the GSEs and jiggering the calculation of portfolio limits to give them more headroom to buy in the delinquent loans. The announcement was a morale booster for debt investors - clearing the way for the GSEs to finance bulk buyouts with debt issues - and removing pressure for portfolio caps meant there was ample room to buy the loans. With those issues resolved, most analysts predicted that the enterprises would move quickly in 2010 to begin buying loans out of pools, sharply but briefly boosting prepayments, particularly in higher coupons (6s and up). Where Is the Surge, Then? Now it's the end of January, and I'm wondering, where is that surge? Loan buyouts are back office operations, generally conducted far from the capital markets, over the automated systems that link GSEs and their servicers, but the debt issues that would fund them are very visible. The enterprises have a nice stash of cash on hand - tens of billions of dollars if my memory serves me - so they can proceed a ways without bringing debt to market. But the give-away indicator mass buyouts were underway would be debt issuance. So far as I could make out from the daily chatter from agency debt analysts, there's been nothing surprising about issuance volume yet this month. And, I've asked a couple of agency debt analysts about signs of activity. They shrugged the question off, telling me to go home and look at prepayments, forgetting perhaps that news of prepayments would lag the buyouts by several weeks. So, imagine my pleasure when I opened the "pdf" from J.P. Morgan Securities' Fixed Income Strategy group this week and read, "Buyout-Watch: keeping an eye on debt issuance." I am not alone in thinking prepayments are not the clue! Although JPM analysts admit that it's possible the GSEs could issue debt after pool factors are announced or fund buyouts by selling dollar rolls, but before settlement date, they recommend watching discount note issuance, because spreads in the sector are tight, "it's a very liquid market market ($100 billion could be issued in a very short period of time), and it matches the short duration nature of a delinquent, high coupon loan." The GSEs, say JPM, could combine issuance of discount notes with that of longer-term paper such as 2-year or 5-year debt. Talking to their colleagues, agency debt strategists, the JPM structured product team notes that so far "there hasn't been much of a pickup," except for a "little bump up" last month. However, "So far in January, we have seen an interesting divergence between Fannie and Freddie net debt issuance, however: while Fannie net debt outstanding shrank by $20 billion thus far, Freddie's has grown by $20 billion." In their report, the JPM analysts put that last point in bolded italics! They say it's too early to draw strong conclusions, but the divergence in funding could imply Freddie is buyout out more efficiently than Fannie at present. That doesn't mean Freddie speeds will be faster than Fannie's - JPM points out that Fannie's overall delinquency rate is about 40% higher than Freddie's. It does mean debt issuance is a "forward-looking buyout signal" well worth watching and JPM is going to be watching it. Me too. 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.