PrimeX Primer; Broker Business Still Impacting Prepayment Speeds; Shouldn't Mortgage Rates be Lower?

(Editor's note: occasionally there is too much news & information to stick with the usual Monday-Friday commentaries. And thus we have today's. Sorry to intrude on any quiet Sunday out there, but what are you doing checking e-mails today anyway?)

At the end of last week a leading Wall Street researcher wrote, "The general tone at ABS East was somewhat gloomy, however most participants view the non-agency sector relatively attractive versus other sectors. The market has seen a recent drop in PrimeX prices, and it is important to understand the collateral underlying in the Prime index and factors that will drive the future performance of the underlying borrowers as well as the cash-index basis." What the heck does that mean?

Practically everyone has heard of the Dow: even a grade-schooler could tell you, "If it goes up, that means the stock market is going up, right?" But it only measures 30 stocks, thus the S&P 500, which includes 470 more, is a better measure but is less quoted in mainstream media for some reason. Does the mortgage-backed security market have equivalents? Yes, one being the "Markit PrimeX," which is a synthetic CDS index referencing a basket of prime mortgage-backed securities. When it was launched its intent was to "to create a liquid, trade-able tool allowing investors to take positions on prime mortgage-backed securities via CDS contracts.  Its liquidity and standardization will allow investors to accurately gauge market sentiment around the asset-class, and to take short or long positions accordingly." So instead of buying $500 million of different MBS securities, one could buy a piece of this synthetic index, and more information can be found here. Or, if you really want to hear more, visit here, and click on the photo of the gal who will give you a lengthy lecture on the whole thing - thanks Josh F. (The internet is an amazing thing.)

These derivatives, somewhat understood by those in the mortgage biz and finance community but not at all by the rest of the world including protesters, have become a bad word. But I bring all this up because these PrimeX derivatives tied to jumbo loans are plummeting in a divergence from the underlying bonds "as firms from TCW Group to Wells Fargo say the credit-default swaps are sending false signals." The prices, per a story in Bloomberg, have "reached record lows this month as trading quadrupled. One index tied to fixed-rate debt fell 10.6 percent through Tuesday, while the underlying bonds declined less than 1 percent, Markit Group Ltd. and JPMorgan Chase & Co. data show. Hedge funds that don't usually trade mortgage debt are piling into PrimeX swaps, seeking the kinds of fortunes that investors earned in 2007 betting against subprime loans, JPMorgan and Barclays Capital analysts said. Trading in PrimeX contracts, whose prices move lower as the cost of protection against defaults on so-called jumbo mortgages rises, soared to $1.2 billion in the first week of October, about the same as in all of September..." When indices diverge from the actual securities to which they are tied, someone typically makes a lot of money and someone is going to lose a lot. "About 12 percent of jumbo mortgages in securities are now at least 60 days delinquent, according to Amherst Securities Group data."

Traders can use them to hedge mortgage production. It is easy to argue against this practice, but traders can also use Treasury securities to hedge mortgage pipelines. But one runs the serious problem of basis risk: one security's price/rate moving while the other does not. During the past week, production coupon 30-year pass-throughs have lagged 5-year and 10-year Treasuries by about .125 in price, but higher coupon 30-year MBS and 15-year MBS have outperformed both their Treasury hedges and lower coupon 30-year MBS over this period. The Fed purchased $5.85 billion agency MBS's over the one week period ending October 19, while domestic bank holdings of agency MBS have increased by $12.9 billion over the week ending October 5.

So mortgages yields have backed up close to 50 basis points from the start of the month, the Fed is buying over $1 billion MBS's per day, the refi index is off 20% from recent highs, and new-production mortgage supply is falling (applications are down). So mortgage rates must be great for borrowers, right?  Not exactly: mortgages have had trouble getting out of their own way of late as volatility has increased, investors are spooked (notice the Halloween reference!) due to the uncertainty, once again, about the government's refi program. Besides, any originator can tell you even if mortgage rates were 1% a good portion of borrowers could not refi anyway. (CoreLogic estimates that about 53% of borrowers with equity in their homes are paying above market rates (defined as the current rate plus 1%, or 5.1%) and higher. About 36% are paying more than 5.5% and 17% are paying more than 6 per cent - think of the refi possibilities!! And about 8 million borrowers who owe more on their homes than they are worth, or about 75 per cent of all "underwater" homeowners, are also paying above-market rates.)

After the Fed announced it would invest early pay-offs back into MBS's, day-trading investors were forced into the market after the mortgage rally and "we now find them a) kicking themselves and b) sitting on the sidelines waiting for better entry points to bring down the cost of their panic purchases." By the various reports I have seen, most folks want to be on the same side as the Fed: buying mortgages.

But mortgage rates for borrowers are running half a percentage point higher than recent historical averages would suggest, complicating Federal Reserve efforts to boost economic growth. Since 2000, rates for 30-year mortgages in the US have generally been about 1.5% higher than yields on 10-year Treasury securities. Earlier this year, the spread between the two rates narrowed further, touching a low of 1.28% in February. A week or two ago, however, the average 30-year fixed-rate mortgage in the US had a rate of 4.18% which is nearly 2.0% higher than the 10-year Treasury yield!

It is one thing to say mortgage rates should be lower; it is another to actually have them there. The spread between US mortgage rates and the 10-year note widened during the summer as investors bought Treasury debt as a safe haven amid growing fears about the European financial crisis. The start of "Operation Twist" in September - in which the Fed buys longer-dated Treasuries to push down long-term interest rates, and then buying mortgages, has caused spreads between mortgage rates and government debt to narrow, but not enough to return the relationship to its customary level.

Lenders say they are charging relatively higher mortgage rates because of tighter lending standards, falling home prices and a lack of capacity to process new home loans, all of which have increased costs. And the Fed can't mandate that, right? Heck, if an underwriter can only get through 2-3 files a day, of course! The number of mortgage brokers has shrunk by two-thirds since 2006. So many companies are keeping the extra spread for themselves. One analyst from Deutsche Bank said, "Higher prices and lower rates on [mortgage-backed securities] only get passed along to consumers at the discretion of mortgage originators, and those originators seem happy keeping rates right where they are." And thus the big lenders don't seem very interested in competing on rate.

Those in the production trenches know that third-party/broker/TPO loans are originated by brokers and correspondents (as opposed to a brick-and-mortar retail arm). And we know that the role of brokers is to advise borrowers, take applications and help select a lender. (Correspondents take the next step as they are able to fund and close a loan in their own name.) For both channels, the servicing is released to the investor. In addition, loan officers are paid solely on commission. As a result, TPO loans are solicited aggressively to refinance at every possible opportunity, and a research piece from Barclays notes the resulting sharp increase in prepayments relative to retail loans when newly in the money.

Should it surprise anyone that broker loans pay off more quickly than retail production? First off, the broker is not dead: a study of recent higher prepayments showed that faster speeds in lower coupon 4/4.5s caught the market by surprise, leading many participants to increase their expectations for new production collateral. However, a closer look shows a strong TPO effect for newer loans (2010-11) which explains the majority of the increase. Barclays recommends "investors find call protection in low TPO% and loan balance pools."

"The October report, in our view, is evidence that the TPO effect is alive and kicking. While the TPO effect fades over time, it could still be a strong prepayment driver in the coming prints. In particular, newer WALA and high balance pools could be more reactive to rates in today's environment. We recommend investors looking to insulate themselves from this effect to consider the following: pools with a low percentage of TPO-originated pools - we recommend investors pay close attention to the TPO%, particularly for new WALA pools. We also recommend that investors pay attention to the loan balances in the pools: We believe the loan size gradient for TPO loans could steepen further from here as the changes to broker compensation filter through the market. As a result, we believe loan balance pools also provide more call protection than in the past. In addition, for lower loan sizes, there is no difference between retail and TPO loans. As a result, investors looking for loan balance and TPO protection for a cheaper pay-up should consider HLB pools."

To sum things up, the good news for brokers is that news of their death is greatly exaggerated. The bad news is that investors are nervous about buying loans originated by them due to the faster-than-average early pay-off risk, and this can certainly impact pricing.

(Parental discretion advised.)

"A cute little girl, missing her two front teeth walks into a pet store. The owner looks at her and says "may I help you"?

The girl looks up and says, "Id wike to buy a bunny wabbitt".

The store owner smiles and says, "Would you wike the pwetty little bwown one or the cute widdle whie one"?

The girl looks at him and says, "I don't think my python gives a thit""

 
If you're interested, visit my twice-a-month blog at the STRATMOR Group web site located at www.stratmorgroup.com . The current blog takes a look at Fannie & Freddie & the FHFA, and the changes they have in the hopper. If you have both the time and inclination, make a comment on what I have written, or on other comments so that folks can learn what's going on out there from the other readers.