We often talk about the month-end trading environment as relatively more supportive for bond markets--all other things being equal.  Reason for this are explained in THIS PRIMER, but here's the gist.  

Let's say you're a money manager who holds a certain amount of bonds in your portfolio.  If you do nothing, those bonds will eventually all be paid off (because bonds are repaid over time), and you'll find yourself with no bonds in your portfolio!  

To remedy this, you'll have to figure out a nominal amount of new bonds to buy every month that keeps your bond holdings where you want them.  "Where you want them" is often determined by a bond market index (such as Barclay's aggregate Treasury index).  It's expressed in terms of duration, which changes slightly based on the total market capitalization of the underlying bonds (i.e. more long-term vs short-term debt being issued would extend the duration relatively more than normal.

Because of changes in the overall composition of the bond market, the index duration doesn't always increase by the same amount, but it always increases (it could theoretically decrease if the government stopped issuing longer term debt and ramped up issuance shorter-term debt).  

But the point is that duration is typically ALWAYS extending because bonds are always being paid down (meaning, as a money manager, you'd have to buy some additional bonds each month to keep your duration in line with the index).   This is the source of the proverbial "month-end bond buying."

Now... traders aren't oblivious to the necessity of these buying needs.  They can do a pretty great job of predicting how much duration they'll need to have added by the end of any given month.  That means month-end buying is only a net benefit to rates when traders have undershot the actual amount of necessary buying.  Whether or not that's the case is always something we have the privilege of observing as it plays out over the last few days of any given month.  If we're to see any this month, it would likely be today, as Monday is typically less liquid in the summertime (fewer counterparties to trade with).

Still, we can guess at how it might play out based on the month's movement.  The risk today and Monday is that the mid-July rally had traders holding too much longer-duration debt, and that there will be some net-selling needs in order to stay in line with month-end indices.  In fact, we almost certainly saw this yesterday in the form of Treasuries' hesitation to follow the "risk-off" move in the afternoon.  So the only question is whether or not that was the last of the net-selling pressure for month-end purposes.

For now, 2.30% has been the intraday resistance level (floor) in 10yr yields, but there are additional layers of resistance just under that.  In the bigger picture, yields need to make a solid run at either 2.22 or 2.40 before they're really saying something  about the next move from the current consolidation.  Risks are slightly tilted toward weakness, but it's still anyone's game.

2017-7-28 open


MBS Pricing Snapshot
Pricing shown below is delayed, please note the timestamp at the bottom. Real time pricing is available via MBS Live.
MBS
FNMA 3.5
102-28 : +0-03
Treasuries
10 YR
2.3049 : -0.0051
Pricing as of 7/28/17 10:03AMEST

Tomorrow's Economic Calendar
Time Event Period Forecast Prior
Friday, Jul 28
8:30 GDP Advance (%)* Q2 2.6 1.4
10:00 U Mich Sentiment Final (ip) Jul 93.1 93.1