MBS RECAP: Calm, Positive Morning Gets Rolled Over By Snowball Selling

"Across all operations in the schedule listed below, the Desk plans to purchase approximately $44 billion and sell approximately $43 billion in Treasury securities over the month of April."

There were some slight changes in the size of the buybacks... 2036-2042 range down from $1.75-2.25 bln to $1.50-2.00 bln. 2020-2022 Buybacks down from $4.50-5.25 bln to $4.25-5.00 bln.

Slightly negative for bond markets here as it decreases the level of guaranteed commitment to the long end, although mitigated by the fact that the overall scheduled buying amount remained unchanged. (In other words, the change is that the Fed is only promising $1.50 bln and $4.25 bln in those two operations whereas before they were in for a quarter of a billion dollars more.

Yet another alert to emphasize the seriousness of the situation. The whole point of the "Tasuki Gap" conversation in this morning's "Day Ahead" post was to introduce the possibility that markets had identified the near term range boundary around 2.14 in 10yr yields after not being able to make it back to the site of the initial gap from March 13th in the mid 2.13's.

It's now possible that we're seeing exactly that. What began as a minor pop driven by some ECB headlines and a few large blocks of sales in TSY Futures quickly took on "snowball" characteristics resulting in the largest 10 minutes of volume of the entire week.

The battle in question is occurring in Treasuries at the moment. MBS are merely keeping pace, and outperforming as the normally do into TSY sell-offs. We're looking for 10yr notes to find support between 2.2068 and 2.2120. If they do this, the bleeding should stop for the day, but it could be too little, too late to escape the sending of the message that the near term range boundary is in. In this case, we'd be looking at an equivocal range trade slightly higher in yield ahead of next week's employment data.

Fannie 3.5's are currently down 6 ticks on the day at 102-24 and 10yr yields are up over 4 bps to 2.2016, having briefly tested the previously mentioned 2.2120 level. Too soon to confirm it as a bounce, but we're hoping.... If you haven't seen a reprice for the worse yet, they're coming. (the tasuki gap post mentioned about is linked here:)

This is more of an addendum to the previous alert to let you know that MBS have indeed broken below 102-29 and despite the "Friday afternoon" light volume in bond markets, price is price, and current prices justify increased risks of negative reprices.

Fannie 3.5's are down 4 ticks on the day now at 102-26. 10's are up 1.8 bps on the day at 2.177.

This isn't necessarily a negative reprice alert, but for our audience members in the field, we wanted to let you know about a quick 4/32nds drop in MBS. 102-02's have quickly become 102-30's and the main culprit looks to be low volume and potentially EU-headlines related to Spain. But like so many snowballs, what starts with low volume can soon grow to problematic sizes.

Bottom line, exercise caution. 10yr yields are up a quick 2bps from their very best to very worst levels of the day in a matter of minutes. It looks like support is at least going to TRY to hold at these levels, which would go a long way toward mitigating any potential shifts in reprice risk, but if Fannie 3.5's dip below 102-29, negative reprice risk would be picking up again.

On the one hand, MBS are at their best levels of the day and have been trading in "tortoise-like" fashion, slowly and steadily making their way higher. The move is analogized by 10yr benchmarks which are also at their best levels of the day. But there are caveats....

a) Low volume
b) day's range is still fairly narrow
c) Fannie 3.5's up 4/32nds to 103-01
d) 10's still above important 2.13 pivot

Long story short, yes, such price action is historically indicative of the early reprice crowd inching their fingers toward to reprice button, but the extent to which that phenomenon plays out is mitigated by the day of the week, hour of the day (afternoon on a Friday = dead), and the general levels.

Whether or not we see positive reprices doesn't much matter next to the fact that we're not likely to see negative reprices (at least not for market-based reasons). To that end, we're floaters until cut-off, unless things turn around markedly before then. Extremely quiet and boring out there this afternoon... We hope it stays that way.