161225 OI tableThe table on the right shows the change in OI over the past “week” (the OI data is 1 day delayed, so from Thursday to Thursday).  The bulk of the move was from the previous Friday (Dec 16), but still, the size of the move was impressive – especially considering how quiet the markets have been.

161225 OI chartIt looks like the “Z algo” was shut down, after being a major player on the curve for a few years.  We no longer see the large kinks in the Z contracts that we had been used to (see chart on left).  It certainly wasn’t a human making these trades manually – the effect on the curve was way too subtle.  It wasn’t completely clear to me what it was doing.  It seemed like it was just taking a large position in the year-end turns for a few years, either as a direct play or indirectly as its curve modeling of the year-end turn was off from the rest of the market.  I also thought it could be some sort of year-end economic model play.  But in any event, the current open interest curve (red line) no longer has the pronounced spikes around the Z contracts that it did just the week before (blue dashed line).

If the 800lb gorilla algo is no longer there, that on the margin makes me feel better about these curve trades we have around EDZ0.  A human is less likely to be systematically sitting on a trade indefinitely – people get into and out of discretionary positions all the time.  We could see a quicker “normalization” of the curve.

Speaking of the Z contracts, we should take a look at what happens to the libor fixing next Thursday…