For bond markets, the week of the Thanksgiving holiday tends to be fairly formulaic.  It's not that the same thing happens every year, but with almost no exceptions, there are common elements and a limited number of permutations.  We began discussing one of those common elements yesterday when I warned about the liquidity problems that tend to characterize the upcoming week.

The simplest definition of liquidity is "volume at price."  That's a bit confusing, so here's a longer definition.  Liquidity is the general level of activity, vibrancy, and depth of any given marketplace, marked by a relatively higher number of buyers and sellers interested in transacting business at any given price.  It is distinct from volume in that a single buyer wanting to buy a billion dollars of oranges at a buck an orange could create high volume, whereas a liquid market would be half a billion buyers and half a billion sellers eager to participate at prices ranging from 50 cents to a buck fifty per orange.   In both examples, the volume is the same.  In the second example, there is infinitely more liquidity.

As you might imagine, the current trading price of oranges in the liquid example would move in a more stable way.  There would be more trades, better price discovery, and almost no volatility, all other things being equal. 

With that in mind, the liquidity issue dovetails into the other typical Thanksgiving week features.  First, market participants know about the liquidity swoon.  They circle the wagons in the previous week, meaning they tend to close more positions than they open.  This persists into the first part of next week, but it does vary by year, depending on the date of the holiday.  The later in the month it falls, the more likely it is that 'month-end' buyers do their buying before the holiday, which easily pushes illiquid markets toward lower yields.  From there, we usually bounce back in the other direction, regardless of the direction of the illiquid run.

In other words, rates tend to consolidate heading into the week, pop in one direction or the other on the week itself, and usually move back in the other direction on the following week.  There are exceptions to this pattern where some of the elements are missing, as can be seen in the 2013 section of the chart below.

This year doesn't necessarily fit the mold of the previous three.  It is a bit disconcerting that we saw a similar move off the lows in 2013 that began in late October followed by efforts to bounce back in mid November.  The big difference is that the selling pressure earlier in 2013 was about 3 times as harsh as that seen earlier this year.  The overall range was also twice as wide. 

All we can do is be ready for the illiquid break that will likely happen, and hope it's in a friendly direction.  If it's in an unfriendly direction, then we have to guard against the possibility that the rest of the year plays out like 2013--at least until such a thing can be ruled out.