There was a curious development where greens sold off 19.4bps last week, but EDM6 sold off… NADA.  Zip.  Zilch.  Really?!?  This was especially strange since I tend to think that if the Fed does not hike in 2016, something may be seriously wrong, so the further hikes in 2017 may not be necessary.  The last time something like this happened (where greens sold off this much in one week and ED2 did less) was October 29, 2014.  This was the time of US Ebola and the PIMCO / Bill Gross unwind aftershocks.  That was not a “normal” week.  And of course being the holder of 8 UNITS (!) of EDJ6 puts (on EDM6) caused me some distress.

Part of this was Dudley being dovish.  But there is probably more to it than that.  Then I started thinking further about something I said last week… that the Fed is not going to surprise hike.  Any hike in the current environment will be somewhat well-telegraphed.  This won’t always be the case, but considering the markets nearly had a meltdown after the December hike, in the near-term I think there is a “zero”[1] chance of a surprise hike.  This should last until the economy and markets look like they are solidly back on firm footing.

Choi CurveThis brings up an interesting dynamic, that I think could explain what is going on with the pricing of near-term hikes.  It is my hypothesis that the current market pricing of any near-term hikes will be somewhat like an “S curve.” (see the chart on the right[2])  The hypothesis goes something like this…

  • If the actual odds of the Fed hiking are near-zero, people will still price something in, so it will rarely be zero. I call this the “lottery ticket effect,” where people play for a big payday even if it is -EV.  Part of this will also be the need to hedge against the improbable scenario.  So what is priced into the market will be higher than the reality.[3]  A current example would be FFJ6, which is pricing in about an 8-10% chance of a hike in March (< 10 days).  I’m not saying a hike couldn’t  It’s just very unlikely, and nowhere near 8-10%.
  • As we move slightly right on the curve (red line on chart), when the odds of a Fed hike are greater but still “unlikely,” eliminating the likelihood of a surprise hike is going to depress the market pricing for the Fed hike. The Fed may think about hiking, but when they see they will take the markets by surprise, they should delay the hike for fear of creating financial instability.
  • As we move further right on the curve, when the Fed hike looks “likely,” the bandwagon jumpers are all going to be piling in. It’s going to be perceived that the markets are giving the Fed the carte blanche to hike and get back to “normal.”  So the market priced odds of a hike are going to be slightly above the actual probability.[4]
  • As the hike gets fully priced, people are not going to risk for example 24bps to get that last 1bp. There will be the “lottery ticket effect” the other way, so the actual hike probability will be greater than the market probability.

In the interests of self-glorification, I will call this the “Choi Curve.”  If Phillips can have a curve with no empirical evidence, why not me?  Sadly, I don’t see how something like this could be proven.  I’m basing this just on my empirical observations and logic.[5]  I think this makes complete sense.  The main disagreement could be the magnitude of this effect near the green dot.  You may think it’s only 1 percentage point (at the widest point), while I think it is closer to 10pps currently.  I think it’s that high because of the current fragile nature of the markets, but I suppose at other times, there could be just a small percentage of this going on.

This has TWO interesting ramifications for profitable trades:

  • If you were bearish, you pretty much want to get in around the green dot in the chart above. Note that the chart is not to scale.  You benefit more from the acceleration in the pricing of rate hike probabilities, and you lose a little less if the rate hikes get priced out.  Another way to say this is that I like being long cheap gamma on some of the 2016 Fed meetings.  I will be the first to admit that the few Fed hikes are not “likely,” but I do think they are underpriced and have sufficient odds of increasing.
  • Unless the markets start pricing in an ease, there will always be a “lottery ticket” demand for Fed meetings, leading into the meeting. For example, I think the odds of a Fed hike in March are less than 5% and could even be less than 1%.  Yet there is much more than that in FFJ6.  So it seems to me, that no matter how unlikely a hike in a Fed meeting currently looks, buying ANY meetings in 2016 for 0.5 to 1 bp makes a lot of sense, if you have ease protection.  Because people love the long shot, and the number of bps in a meeting should stay positive (assuming no eases).  People also need hedges in case of an unexpected blow-up, which could cause them to pay up.

This is why I still like the EDJ6 99.25 puts vs FFK6.  FFJ6, which is imbedded in FFK6 (= March + April meetings), needs to give me my 2+bps in 9 days when the March meeting lottery ends in a bust.  FFK6 should rally in sympathy.  But if it doesn’t rally, that means the slope in this part of the curve has steepened noticeably, and we are long the June and 50% of the July meeting in the puts.  The latter two meetings are much higher probability in my mind.  So after the skip in March, this trade is one April meeting (think Dudley frowning) against one June meeting AND a half a July meeting.  I’ll take the latter.

This is also my reason for liking trades F11 and F11b, both of which involved buying FFN-Q spread for 1 (and below).  However, there are other related plays on the curve.  Look for a new trade write-up (Trade F13) early next week, as well as the write-up for F12 that I emailed Friday.  I think we may get some pre-ECB rally / profit-taking early next week, so Wednesday / Thursday may be a good time to start thinking about adding to “smart” shorts ahead of the FOMC.  I have a preference to do some before the ECB and some more after.

I will probably review Fed pricing next week and discuss some thoughts about the meeting.  The executive summary is that I lean bearish for the Fed meeting.  Employment and inflation have been firmer, the markets are more stable, and oil has recovered some.  There’s a chance that the “closely monitoring global economic and financial developments” phrase may be weakened, and there is a chance the “balance of risks” statement could be put back in.  It seems foolish to insert a new weaker balance of risks phrase (even if they believed it), so I think they just leave it out, or put back “balanced.”  These things all lean slightly bearish in my mind.  That January statement was the one that had the real potential for dovishness, and things look much better since then.

[1] As close to zero without it actually being zero.

[2] Note that the chart is only illustrative and not to scale.  Do not make any conclusions based on the numbers (where the lines cross, amplitude, etc).

[3] I had previously mentioned my undergrad thesis advisor’s research on how the favorites in horse racing are generally undervalued because people like betting on the long shot.

[4] I think the market pricing differential when a hike looks “likely” will be less than when a hike looks “unlikely” because maintaining the status quo is probably easier than changing it.

[5] And making excuses for why EDM did not sell off more last week.   Eh hem.