After Friday's impressive (and much-needed) bounce toward lower yields, the new week/month took on even greater importance than normal as a venue for a shift in the recently ultra-bearish tone.  In other words, yields have been surging relentlessly higher, and it's about damn time for a ceiling bounce (or so we'd hope). Friday's afternoon gains increased those hopes, and maybe even the odds.

If we get the bounce we're hoping for, you'll know it.  It will be big and decisive.  If, on the other hand, yields merely chop around in a wide, sideways range (let's say 1.38 - 1.62), it will be confirmation that the market is entering the "acceptance" phase of it's grieving process over the loss of low yields.

Why focus on the levels of 1.38-1.62?  Chart nerds already know where I'm going with this.  38% and 62% are fibonacci numbers.  Market technicians look for support/resistance when a chart has recovered or "retraced" certain percentages of any big rally or sell-off (or simply any prolonged movement from some specific point of origin).  For instance, we could set a start point of the high yields in late 2018 and a low point of the all-time lows in early 2020.  The fibonacci followers are expecting the yields that line up with the fibonacci sequence to be more consequential than other yields.  Here's where they fall over that time horizon:

20200301 open.png

That trading range is so large that it really doesn't offer more than a few fibonacci options.  The first (at a yield of 1.0% or 23.6% retracement) indeed proved to be important.  But it was at 1.0%, which is also a key psychological level, so it's hard to chalk the bounces up to fibo retracements.  1.43/1.44 has been on our list for a while, largely due to the big bounce in 2019 (no way to know at the time that it would end up as the 38% fibo retracement, but it did).  

But again, this trading range is too wide to get many cues from retracement levels.  Since the market has moved so much in a short time frame, how can we set the boundaries to yield more retracement options? Of course we just shrink the boundaries such that the central fibo retracements coincide with most of the recent trading, but we need to come up with an objective and repeatable rule to inform the shrinkage.  One of the most objective ways to do this is by simply using big, round integer levels as the 0% and 100% retracements.  In other words, we can line up 1.0% 10yr yields with the 0% fibonacci line and 2.0% with the 100% fibo line.

The result is the sort of thing that tricks impressionable market watchers into falling too far in love with technical analysis.  The huge opening gap in mid Feb?  Right on the 23% fibo.  The first major ceiling battle and the massive breakout that followed?  The 38% retracement was the 50 yard line for all that.  The highest intraday yields of the move?  Perfectly aligned with the 62% fibo line.  The stuff looks like magic in this context.

20210301 open2.png

But is it magic? 

Yes and no.  It's not nearly as magical as it looks.  Things have panned out very well for this particular run, assuming we're all OK arbitrarily choosing 1.0% and 2.0% as the retracement boundaries.  And in fairness, some traders and trading algorithms are no-doubt paying attention to these pivot points.  Whether or not they're doing so in sufficient volume to dictate momentum is a different story.  Could it contribute?  Sure.  Could other traders with other motivations cause bigger bounces at other levels?  Most certainly. 

Bottom line, if you'd like to watch 1.38 as an important line in the sand because it's a fibonacci level, have at it!  We'll be watching it simply because it proved to be significant on the way up (and because it's in a significant historical zone with the 1.3's marking the 2 previous runs at all-time lows in 2012 and 2016). 

Moving on to MBS, we have a few housekeeping tasks to start the week.  To answer the very common question recently: 100% of your focus should be on the UMBS 2.5 coupon.  It's the only game in town as far as rate sheet relevance is concerned.  Here's why.

Also keep in mind that MBS price swings have been and will continue to be large from moment to moment (depending on the moment in question).  This can give the appearance of big swings in prices even when that's not actually where trades are occuring (remember, there is a BID price, which appears on MBS Live, and an ASK price which does not.  If bid and ask are far apart, the true market price is a bit of a moving target until buyers and sellers get back on the same page).  In any event, we will always give you a heads up if price swings are being unduly distorted by illiquidity.  

Important econ data on tap this week: ISM Manu and Non-Manu, ADP, NFP Friday. 

Away from econ data, Fed Chair Powell speaks on Thursday.  The market would like Powell to acknowledge that high yields (and illiquidity) could eventually be a concern, as it was in March 2020 before the Fed unleashed it's most aggressive ramp up in bond buying, ever.