For bond markets, today has largely been about holding ground and retrenching.  The most relevant example came during the overnight session--just before the domestic open--when 10yr yields bounced convincingly off the 1.84 pivot point.  Keep in mind that this is one of the most important long-term inflection points for domestic interest rates.  Since breaking below last Wednesday, it's seen a lot of action and most of it supportive.

The resilience is also being supported by a bounce lower in equities and oil.  Perhaps a better way to say this would be that equities, oil, and bond yields are moving in unison, away from risk.

On an MBS-specific note, things are a little less awesome.  Volatility always takes a toll on MBS' ability to follow Treasuries to lower yields and today is no different.  A compounding issue is the 'flattening' bias in the yield curve.  This means that longer duration Treasuries (like 10's and 30's) are gaining much more ground than shorter durations (like 2's and 3's).  Although the calculations are subjective, most market participants would agree that the average mortgage in a Fannie 3.0 pool is nowhere close to having a 10yr life expectancy.  In other words, it has a shorter duration.  And since we already know that shorter duration debt is underperforming today, we have yet another reason for MBS to be lagging the 10yr move.

Back to the topic of volatility now...  Perhaps more insidious than its effect on MBS prices is its effect on lenders' ability to adjust pricing to match MBS moves.  This is a very simple concept when you get down to it.  The wider the range of potential near-term movement, the more expensive it becomes for mortgage lenders to do business.  If the market goes one direction, lock fallout can increase.  If it goes the other direction, lenders can have a hard time getting a high enough price for loans that locked at lower rates (i.e. higher prices).  Of course, much of this depends on a lender's specific hedging strategy, but the fact remains that hedging, in general, becomes more expensive as volatility increases.  And so it is that a big move in the opposite direction from the previous move isn't having the expected impact.