Posted
Damage Report 5/27/09
The dust is finally settling and ashes finally floating
away, at least for today. We are left
with one of the most grotesque images to come out of the MBS marketplace in
years: the visual representation of the deservedly named "Black Wednesday."
When all was said and done, this was over a TWO POINT DROP
on 4.0's (66 ticks to be exact) which forced us to shift analysis to the more
current coupon (as of today!): 4.5's which "only" lost 1-20 (52 ticks). Indeed, the duration shedding we've been
discussing was evident in the drastic outperformance of the higher
coupons.
Plain and Simple: this means
that many fixed income accounts are selling debt that has a long maturity as
economic prospects improve. Lower rate
MBS is longer maturity (since mortgage holders are less likely to refi) and
corresponds more to the "longer" end of the treasury curve (5-10 yrs) which
also got crushed today.
But unlike days past where MBS played the tortoise to Tsy's
hare and executed mimetic movements on a smaller scale, today was tsy's day to
answer to MBS. To say MBS were center
stage today would be putting it mildly.
First of all, we need to remember that for years now, total outstanding
MBS has far surpassed total outstanding tsy debt. So when something big happens in the MBS
market, it can absolutely send shockwaves throughout the entire fixed income
sector. We'll elaborate more later in
this post and tomorrow, but today was the story of an MBS snowball that
consumed everything in its path, including tsy's, until it reached such a
momentum that only the bottom of the mountain was a sufficient backstop. Thinking of the "mountain" as literally ALL
the gains we've had since the Fed started buying, you can see this is a
not-unreasonable metaphor given the day over day chart:
It's a cruel peak indeed that surprises the complacent skier
with such a sheer drop-off. Who knew
just how important it would be to cross the annual midpoint, the 50%
retracement support that has been a constant topic of conversation on this
blog! Market technicals are uncommonly germane
at the moment.
Is this the worst or one of the worst days ever?
The rarely seen week over week candle chart
shows how this week compares to the worst weeks in the past 12 months:
The solid candles, for those not familiar, are weeks where prices drop. Right away our eyes are drawn to October when the credit crisis kicked into high gear. 10/24 was the only single day in the last 12 months that 4.5's lost more than today 1-25 versus 1-20. The week of 10/12 was also worse than this week, moving down 2-31 versus 2-20 so far this week (2 days left though!). After that, there are some close runners up with a 2-12 and 2-14 weekly loss in June and July respectively. All in the same nasty, horrible ballpark. The weekly chart also offers some technical guidance. The small candle with the long shadow (that's the "wick") on the far left side of the gray line is technically significant in a couple different ways. To discuss them in detail would be a post unto itself, so suffice it to day that it is important that the selling ended where it did today. If we go no lower tomorrow, the 99-10 price level will start to prove itself as more reliable support.Whatever the case, if we had to pick one point below February's lows as "the bottom of the mountain" that candlestick would indeed be the week. So much is at risk given the fundamental data injections that will take place starting tomorrow, but it's not naive to hope that this floor will hold, even if we dance around it in the coming days. Furthermore, we will see a return of the normal "MBS acting according to tsy cues" as opposed to today's opposite.
The Why: Technical Logic
But WHY did MBS bring tsy's down? Why all the selling? Why today?
Why all at once? Why only
technical warning signs and limited fundamentals? Why why why?!
For those of you versed enough in "spreads" and a smattering
of trading savvy, the following chart does what it says in a general sense, in
that it explains today's selling in a single chart. In short order, all of the fundamental considerations (to be discussed) fueling the graphical representation of recent market behavior...
This chart is really the visual of the whole enchilada. We're looking at "spreads," not MBS prices. For the unaquainted, this line is the difference between the yield of a 4.5% MBS coupon and a 5yr US Treasury (5UST). MBS yield involves calculations based on prepayment models and the yields involved in this spread calculation are based on a fairly standard prepayment model that is very much in line with others. Once prepayments are known and factored into the model, a yield can be calculated which is usually somewhat similar to the MBS current coupon. That yield is subtracted from the 5UST yield and we're left with spread. Check the "speeds and spreads" for more info.
Much much much more than the actual PRICE of MBS, Spread is the main consideration when it comes to valuing MBS. This is "relative value." In simple terms, MBS may be worth 101-27, but relative to what?! By putting it against a risk free benchmark with no embedded call option such as a treasury, one gets a much better idea of the value of MBS. The lower or "tighter" the spread, the more relative value, the higher or "wider" the spread, the more relative cheapness (AQ and I ranted endlessly in November about what an insane bargain MBS was RELATIVE to tsy's).
Basically, by looking at the chart, we can see the spread improving steadily and with a narrowing trading range ever since the Fed announced buying in late Q4 2008. Anyone who says the Fed hasn't accomplished anything with this should be reconsidering that statement upon viewing the above chart. In the middle of last week, spreads were at the lowest point in over a year, a "too tight" zone. This was occuring even as treasuries began to sell and yields moved higher, even then we witnessed MBS coupons outperform their benchmarks.
Looking further into the yield spread behavior illustated by the above chart. Going back to the last major bottom in Mid August '08, we have a clearly violated double bottom (technically speaking). Look what happened just after that! Spreads widened quickly! Relative value was so rich that MBS investors could readily cash in, having waited as long as they thought they could before dumping the supply. To further edify the technical explanation look at the little "hitch" in the curve as it crosses that same point! It actually bounces back up off that exact point experiencing the resistance of that implied floor that just caused a bit too aggressive of a widening spree. But the markets decided the waters could be tested to yet tighter levels for MBS. We crossed that line and found ourselves up against the annual "tight" we'd created just two weeks earlier. If we broke that line, it can be traced all the way back to yet another low from April 2008. From a technical perspective, this was the trigger that coincided with all hell breaking loose today, and unleashing the previously discussed snowball...
MBS 5pm "Going Out" Marks
Special Report on MBS "Black Wednesday: Part II to come...
Secondary Marketing Managers and Capital Markets Desks, if you are interested in subscribing to the same fixed income and mortgage market data we use:
.