After missing out on the Asian bid overnight and facing potential concerns among yesterday's buyers with a potential classic case of "buyer's remorse" MBS have been fighting this morning to hold yesterday's gains.

Current coupons are straddling 5.0's and 5.5's. Both of those stacks had sold off quite a bit about an hour ago. But as stocks have sold off and as servicer and originator pipelines take positions for the impending 48 hour Class A settlement, we're seeing MBS push back to unchanged day over day.

5.0's are at 99-13 and 5.5's are at 101-05 FOR THE OCTOBER delivery. Now HERE'S SOMETHING YOU WON'T likely see discussed on the "other" MBS analysis websites. This week is a great example of a time where we have to look at the following month delivery prices.

WHY?!

As we have discussed, 48 hour Class A settlements (which refers to the 48 hour, 2 day period where Class A which includes Fannie and Freddie 30 yr MBS, occur right around the 10th of each month. We have class A settlements coming up on Thursday. This creates an artificial "overnight" drop in MBS price to the next month's coupon prices. Now, we normally gradually see those prices recoup the different between last month's prices and the current month (in this case Sep to Oct).

In layperson's terms this means that the well priced 5.0 september coupons, which are at 99-18, will basically turn into October coupone more or less overnight. Under normal circumstances we see trailing month (October in this case) coupons lag current prices by about 8-9 ticks. Indeed, this is the case in 5.5's. But in 5.0's there is only a 3 tick gap. We'll revisit that in a moment, but the point is, as soon as that settlement occurs the currently traded coupon will become October, and as we discussed, the trailing month will almost always carry a lower price. This LOOKS like MBS loses day over day. But what we normally see, all other things being equal, is that as MBS production shifts to current month coupons, prices will rise back up to the previous month's levels.

Here's the beautiful thing. September appears to be more or less "in the books" at current levels. We are not seeing very many trades at all on September coupons. Servicers, money managers, banks, etc.... appear to be 'full up' on that paper and are shifting their trading action to October. What this means is that we may have a rough day or two as lenders are forced to price their sheets based on worst case scenario, especially the ma and pa correspondent lenders that might not be considering some of the more esoteric factors we are discussing here.

Moreover, your average consumer of MBS data reading you average MBS site, could well be left wondering "why did we just lost 4-8 ticks overnight for no good reason? At the very least, now you won't need to be wondering why that occurs, but something even nicer is occuring. The october and november coupons are tightening to current coupon prices, especially in the lower end of the stack (4.5's to 5.5's). This is actually a GREAT piece of secondary market information/psychology. Basically, we are seeing MBS traders keep making that "down in coupon" bet. So the October delivery will be more important for us to track currently as it should be a better guage of where the biggest lenders will be setting rates as it's highly unlikely that they will be trying to squeeze much more into the jam packed September delivery.

ON TOP OF THAT, after a few conversations I've had with secondary desks at a few of my lenders, I'm being told they are intentionally pushing production out for a Novemeber delivery! For instance, on an example loan, I asked a lender if we could still submit DPA as long as it closed by 9/30 (this was the conversation that prompted and kicked off my research). The AE, knowing that I'm interested in such things, did me the courtesy of putting me in touch with their price desk. They informed me that they would take the submission on an exception ON THE CONDITION that it COULD NOT sign and fund in September. Huh?

In a nutshell, they did not "have room" for any more of those deals in their October delivery. So now to the thesis. Not only is the September settlement already 'sold out,' but as a result of what have largely been improving rates so far this month in conjunction with what was an "all time" lock day for many lenders yesterday, their forward commitments for October deliveries have become "sold out" much more quickly than normal. So what they are doing is "pushing out" current submissions to sign in October as opposed to September, thus making those loans intended to go into pools that are settled in November.

Why is all this important? Bottom line, October and November settlements are much tighter to September settlements than they normally would be for the two aforementioned reasons of lower rates in general combined with a more than immense smattering of locks yesterday. AND if we can see that traders are buying "down in coupon" (i.e. the lower end of the stack is not only better bid, but also has a narrower difference between settlements) it clues us in to market sentiment. What's the sentiment? In a word, positive.

Asia is waiting for us to prove the soundness of MBS, and they were also hesitant to buy at what they perceived to be an "exuberance premium" last night. They've heard so much talk about "bailout" for almost 2 months now with little decisive action. As yesterday was the first decisive action, AND since it coincided with a dramatic run up in dollar price, one might safely assume that Asia would not quite yet be enticed to get on board. But we hope they soon will. The rest of the buy bid is feeling quite tired after already riding the "tightener' this month. Money managers are tapped out, banks have had to sell into Par coupons (remember banks can't hold anything over PAR), "real money" (insurance funds, etc...) are similarly satiated, not to mention needing to see a bit more stability before coming to terms with MBS risk, and "fast money" (hedge funds, etc...) are now blocked out from the "relative value" ("hey! that's underpriced!") bid that they so dearly love (because MBS have picked up the bid to a greater extent this month, relative value has decreased as spreads have narrowed).

And with respect to all of the above, a one day announcement that spikes dollar prices by about a point and half, that ALSO coincides with about a 60 tick tightener to the basis is all "a bit sudden" for the aforementioned market participants to digest. It's as if the underdog team swung for the fences yesterday and hit it out of the park, they might all stand around looking at each other with those expressions of "did that just really happen? what next?" It might take our bad news bears a moment or two to realize that it's safe for them to get back out there and swing the bat again.

So, we can assume the normal nominal price gap between current coupons and the trailing settlement month and keep a closer eye on those October delivery dates, as well as closely monitoring the 'down in coupon' buying we are currently seeing. If, under normal circumstances, we see a 7-9 tick gap between settlements, (although the following is an oversimplification), we can apply some simple algebra to derive where current coupons "should be" in light of everything else we've discussed.

SO! Ipso, ergo, therefore, etc... because the gap between settlement dates is a bit narrower than normal, and because those gaps tighten as we move down in coupon (i.e. 4.5's are only a 4 tick gap whereas 5.5's are a 8-9 tick gap), we are left with an ever-so-slightly positive sentiment for MBS. Furthermore, word on the street is that the Fed wants to have a more convincing tightener than what we saw yesterday. With spreads gapping out slightly again today (buyer's remorse?), this is also a good sign, even though no one knows exactly where that current nominal spread would be assuming a data vacuum.

Whatever the case, the take away is "it's not quite as bad as the day over day price changes would indicate. Although lenders will have to price conservatively, assuming a no-impact for the Thursday and Friday data, we should gain and tighten due to intrinsic MBS forces. Even so, current coupons are at 101-11 which is right near the bottom of yesterday's high range. You'll see ratesheets today either an eighth or a quarter worse. Things will continue to fluctuate as bonds follow stocks and MBS follow Bonds today, but at the end of the day, if your lender has squeezed more than a quarter out of you, keep floating, even if you are a short termer. If there's evidence to suggest those rates will hold, you may consider locking tonight, but ONCE AGAIN, the mid term sentiment of the 'float club" remains: the more time you have between now and locking, the safer and safer the bet becomes to keep floating.

Now, though, we have even more fundamental data to support that position. The ONLY thing we have to closely watch and assess is the determination to what degree yesterday's run-up in MBS was over-exuberant. But considering the laws of MBS nature, it wouldn't be quite right for MBS to pack on extra gains after such a huge day yesterday. Just know this: our government has made no pretense about its intention to attract foreign buyers to MBS. Whatever occurred yesterday certainly did not do that! SO, what would you conclude? If the "wait and see" approach doesn't do it, more MBS-supportive actions are to come. And don't forget Uncle Same has 100 billion set aside to pump into Frannie, and yesterday was only the first 2 billion.

My friends, my visceral instinct says they have, at least to some degree "figured it out" inasmuch as they believe lower mortgage rates will stimulate home-buying, halt price depreciation, and mitigate the impetus of non-distressed buyers to walk away in foreclosure, bankruptcy, or short sales. We've been discussing "retracement" as a common theme historically during these fast run-ups. In the absence of MBS headline risk, an unexpectedly boomy economic picture from this week's data, or a ridiculous inflation headline on friday, the positivity in MBS is more likely to continue after we get through this one step back. Here's the "two steps forward!"

(short term lockers: your course of action depends largely on loan type. Treasury ARMs may be at a price level that justifies locking, but Fannie/Freddie MBS will be all about Thursday and Friday's data after we get through the "position shuffling" of today. I don't offer ________ strategy, where the blank is filled in with a scheduled economic release. The largest financial firms in the world employ analysts and economists whose primary function is to predict/estimate the results of these scheduled reports. For me, or anyone else to presume to do it better would be either megalomania of the worst order, or a delusional claim to understand market forces far beyond one's station (assuming the one doing the delusional claiming was not him/herself a successful economist/analyst. We can never know where these reports will come out. So always consider that potential 2 way volatility if you have short term locks).

The rest of you, see you at float club!