Month/quarter-end buying was an important part of this week's strength through Wednesday. It served as sort of a constant, mildly positive force helping keep yields anchored despite an attempt at a bounce in equities markets.

There's no rule that says bonds had to follow stocks higher yesterday, so we shouldn't be overly-surprised that they didn't. Still, there has been quite a bit of that 'stock lever' correlation lately--enough that it makes sense to figure out what changed--if anything--in order to bring about the new market behavior. 

We can definitely chalk the change up to the month-end trading environment.  That might seem to imply that we're heading into the rest of the week at a disadvantage, but it doesn't really work like that.  The prevalent tradeflows heading into a new month will be whatever they will be.  If they take bonds back into weaker territory, it won't necessarily have been because month-end trading was a temporary boost.  After all, there are opportunistic traders that are fully capable of taking advantage of the situation if it was that obvious.

Moreover, no one can ever be fully availed of all the forces acting on markets at any given time.  It's one thing for us to know that there was 'month-end buying,' and that it had a positive effect on bonds so far this week.  It's another thing altogether to know whether or not there were 47 other positive and negative considerations that aligned in just such a way to make it look like month-end buying was leading the strength. 

We can come back to the stock lever here to conclude the thesis.  It's one thing to explain bond market movement in the context of equities markets over the time frame in the top chart.  It's another thing altogether to try to do that with the bottom chart:

Bottom line, there's not much sense in getting overly analytical about the stuff in the white circled areas above, especially with NFP coming up tomorrow. As for today, we have several pieces of economic data, but only one significant report.  That's ISM Manufacturing at 10am.  Any big beat/miss has the potential to cause a fair amount of movement.  Even without data, it's not uncommon for bond markets to begin taking a 'lead-off' ahead of big-ticket data, much like we saw a few weeks ago heading into the FOMC Announcement.