Why is December( and in a lesser extent November ) a little richer than other months every year on the curve? I thought it had to do with year end taxes on inventories, but in this case if refiners stop accumulating in October and starts back buying in January, the effect would be less demand in dec and more in Jan, hence higher price in Jan... What am I missing? Thanks.
Here's a guess - and it's just a guess: Index funds have a lot of different stratgies to minimize transacation/roll costs, and I believe that one strategy is to just buy the December contract and roll to the next December contract. December seems to have the most liquidity of all the months. So, to the extent they are net buyers, the December contracts will be bid up relative to the other 11 months. Of course, I could be completely wrong.
Not stupid at all. I didn't think about the fact that December drives more volume than other expiries. It makes sense particularly for further years, where there is no volume and settlements for other expiries are based on other thing than actual last minutes prints.
It's not a seasonal pattern as if Dec was bid up against other expiries going into spring , or fall or other periods. It seems permanently overvalued. In contango, the curve looks a little bit exponential and in an inverted markets, it goes "flat" toward the end of the year. I will try to post a chart tonight.
The front 12 months have been in contango for the last ~12 months or so IIRC, ever since the Iranian oil embargo was announced. Z13 is just the current peak. The curve keeps rolling over the peak as old contracts roll off the board. October, 2011: February 1, 2012: February 22, 2012: April 4, 2012: September, 2012:
Yes I know for Dec 13. But look at Dec14 and Dec15, there is a slight strength for each on the declining curve. I had noticed this before, one or two years ago when the curve was quite different. Thanks for the charts.
Can you post an example? My initial thought is that you're seeing phantom up/down ticks. The Dec/June contracts are the most liquid in the backs, so they will trade occasionally, but the other months will not, so what you are seeing might just be stale quotes on the non-Dec/June bullets.
It can be seen pretty clearly on your April, 4 chart. Dec14 and Dec15 are easy to notice on the curve. However, I guess you are right. It's just stale quotes or if your charts use settlements, a difference in the way settlements are calculated depending on whether or not the expiry traded during the day.At first I thought it was a real pattern ...
Yea, my curves are updated real time, but only for the liquid months. All the non-Z/M bullets out more than 18 months are typically stale settles.