Trading activity was fairly decent in bond markets today--especially considering the seasonal norms, though this is largely a factor of Thanksgiving falling on the 26th.  That means that there will only be one more full trading day in November, and participants aren't expecting it to be awash in liquidity.  Combined with favorable month-end trading demand, it's fair to assume that the boost in volume is due in large part to early month-end trading.  Then there's the matter of next week's significant events (ECB Meeting and NFP) that are likely increasing traders' urges to be positioned appropriately.

Volume is one thing.  Market movement is another.  We're just not seeing much movement this week, suggesting deeply entrenched positions on either side of a narrow battlefield.  The center of that field continues to be something around 2.25 in terms of 10yr yields, although the real midpoint is a diagonal line bringing short term and long term yields closer together.  A bit abstract for the day before Thanksgiving, I know, but the point is that bond markets aren't as simple as a mere 10yr yield even though we often discuss them as such.  The yield curve has been flattening steadily for the past 2 weeks, and it's just so happened that 10's haven't had to move much in order to facilitate that.

This focus on the relationship between various parts of the yield curve as opposed to outright levels is known as the 'curve trade' among other things.  The fact that longer-dated Treasuries can stay flat also implies flatness for MBS, as the typical mortgage lasts between 5 and 10 years.  Certainly though, we can see the preference for that longer-term debt in the MBS underperformance of 10 and 30yr bonds today.  It's especially noticeable when looking at the moderate weakness in Fannie 4.0s (which have a shorter life span than 3s or 3.5s) compared to the modest gains in 10 and 30yr Treasuries.