There have been reasons to fear a momentum shift in bonds for several weeks now.  Here's how we discussed it at the beginning of October:

"It's never a bad idea to consider risks on the road ahead--especially when things start deteriorating at the beginning of the month. We often see a shift in momentum with a new month when the previous one was fairly consistent with a certain theme. September's theme was definitely consistent."

We went on to discuss a small scale breakout of a consolidation pattern that occurred well inside the already super narrow .62-.72 range. At the time, we concluded that things would be getting more serious when the .72 ceiling gave way.  4 days later, it did.  The entirety of the 10/5 - 10/9 week was spent bouncing at the 0.79% technical ceiling.  That left us in a similar position compared to the showdown with the 0.73% ceiling--another bridge we'd cross when/if we came to it.

As of the overnight trading session, we've come to it!  Yields were as high as .836% briefly, which is easily enough to consider the range boundary to be broken.  Going back to that theme of a new month bringing new momentum, October's 2nd major range breakout (.79%) hurts even more than the first (.72%).

20201021 open1.png

However painful this may be in the short term, the move is best described as gentle in the longer term.  In fact, in the bigger picture yield range of 0.5 to 0.96, one might even say we're moving logically from one boundary to another.  In exactly the same way that we concluded a break above 0.72 would be a relevant tipping point for a short term range, it would take a break above 0.96% to send more dire messages about the longer-term range.

20201021 open2.png

Revisionist history be damned.  This is nothing new.  Throughout the weeks spent inside the 0.62-0.72 range, the MBS Live Huddle (Market Summary) had the following in the "lock/float considerations" section:

In the bigger picture, there's a wider sideways range to consider marked by 10yr yields between 0.5 and 0.95. When the Fed talks about low rates for a long time, this is the type of range they have in mind for longer-term rates. Bottom line, we may see pockets of much narrower range trading, but those ranges are destined to break much more readily than their wider counterparts (the 0.62-0.72 range was an anomaly in that regard).

It would be really nice to know if the more dire ceilings are destined to break well before we get there, but we can't really know.  The pandemic makes past precedent a less reliable predictor of future behavior compared to other bond market mega rallies.  That said, there is no historical precedent that suggests yields will remain under 1% forever, even if a day comes in the future where 1%+ yields are fully extinct. 

None of us know exactly when that will happen (the rise above 1%, not the extinction...).  It might take weeks, and come in response to the election/stimulus.  Or it might take months and happen only slowly in response to evolving pandemic/economic conditions.  Either way, it's a good eventuality to be prepared for.  It's up to you whether that means being more risk averse as yields move up, or being more prepared to capitalize on the rebound when/if yields move back down.

Fortunately, MBS continue to defy the odds with respect to outperformance vs Treasuries.  Compounding the pleasant issue is the fact that mortgage rates are so much higher than MBS suggest they should be that lenders really don't have to hit rate sheets much at all when we see these sorts of Treasury sell-offs. 

Just be aware that if a sell-off gets big enough, it can lead the mortgage market to question the viability of lower MBS coupons.  This is known as "extension risk" in the secondary market.  It means investors don't want to be caught holding MBS that will never refi at rates far lower than prevailing bond market rates.  When those fears ramp up, that's when we can see the most aggressive selling in MBS, and clear departure from the recent trend of outperformance.