The Return Of The Year-End Eurodollar Futures Turn

by: Curve Advisor


The 3mo LIBOR fixing has increased 19 basis points in the past three weeks, in part due to year-end funding pressures.

There is no reason to think we wouldn’t get some funding pressure in future years.

Year-end funding pressures are not priced in further out the Eurodollar futures curve.

Even if you don’t want to do a specific year-end turn trade, you should consider checking your EDZ exposure up and down the curve, to make sure you are not overly long the EDZs.

Have you ever heard that phrase “Many traders have never seen___”? For example:

  • Many traders have never seen a real vol spike. You’ve probably heard ad nauseum about how short vol (VIX) sellers will get crushed one day.
  • Many traders have never seen a real equity correction. For almost eight years, buying on dips has been a good strategy.
  • Many traders have never seen buying fixed income against the bond (or tens) yield downtrend line in the US not work. There are many old traders who have made a career on having the 30+ year downtrend line in the bonds hold.
  • Many traders have never seen an inflation spike. Inflation has been fairly tame the past two decades.

To that list, you can now add:

  • Many traders have never seen real year-end funding pressure, aka the year-end turn.

The year-end turn in Eurodollars has been something that I first read about in Burghardt’s classic “the Eurodollar Futures and Options Handbook.” For those of you who want a primer, here is a link to a five page article that discusses the year-end turn. Back when I first started trading in 2000, the year-end turn was about 6 basis points. In other words, all the EDZ contracts were about 6 bps cheap to a “smoothed” ED curve. This is because the markets expected 6bps of funding pressure over year-end. Since EDs are forwards, it made sense to price this in over the entire curve.

The chart below shows a rough estimate of the year-end turn priced into the EDZ contracts in the blues and golds (roughly the 1 year rate 3 years forward and one year rate 4 years forward). I thought the turns further out would give a cleaner look, since they are mostly free from the noise of any hiking and easing cycles. In late 2002 (post the dot-com recession), the year-end turn collapsed and didn’t come back on the recovery a few years later. Then we had the LIBOR crisis in 2008 and the turn came back. When LIBOR explodes, it is understandable that the markets would be especially concerned about funding around year-end. As we recovered from the Great Recession, we went back to having minimal year-end turn priced into the Z contracts.

Source: Curve Advisor Source: Curve Advisor

This year the 3mo LIBOR fixing has exploded from 1.49463 on December 1 to 1.68577 on December 22. Only 1bp of the rise was attributable to an increase in the December Fed meeting. Another 1bp is attributable to the March meeting being included in the latest 3mo LIBOR fixings. That means that we had a net increase of 17bps in the spread between 3mo LIBOR and Fed Funds in three weeks (=1.686-1.495 -0.01-0.01). That is an insane move in a low volatility, yield-grab world. Last year, 3mo LIBOR only rose 5.5 basis points between December 1 and December 22, and some of that was again due to a December (2016) hike and a smaller amount of March (2017) hike being priced in. So 3mo LIBOR rose only 4bps last year on net over Fed funds. That is still more than the one basis point that is typically priced in, but perhaps not enough for the markets to take notice. The whopping 17bp increase this year is hard to miss. Source: Curve Advisor Source: Curve Advisor

I have this theory that the turn should increase as rates increase. The chart above shows the year-end turn as a function of the level of rates (I chose the rate of the ED red pack to be on the x axis). We can see a clear increasing relationship in both the pre-2008 and post-2008 dots.

There could be many causes of this year’s LIBOR increase. We have some pressure from regulatory capital requirements, possible repatriation causes (post-tax-reform money waiting to be sent), tight funding in general (Treasuries special in repo), possibly something related to Brexit, possibly something related to a budget impasse, etc. The tapering of the Fed’s QE could also be putting upward pressure on year-end funding. The Fed is taking a little bit of the punchbowl out every quarter, after all, that could add up over time. While there could be some one-time factors, it is likely that at least some of the factors that caused LIBOR to widen this year could appear again next year… or every year! For example, do we have any reason to think banks won’t be hoarding cash next year for regulatory reasons? They can plan ahead better, but that may mean LIBOR just starts increasing earlier.

Since the Eurodollar futures are forwards, we should expect some turn premium to be priced into the Z contracts further out the curve. As you can see from the charts above, the year-end turn is going to be positive. We just witnessed a 17bp spike in 3mo LIBOR. Call me a simpleton, but if I see something move 17bps unexpectedly, and part of the reason could be an annual phenomenon, AND there is nothing priced, I’m going to think of it as a “free option.” If it can’t go down, maybe it’ll go up.

It has been a little perplexing how the markets have been slow to adjust the rest of the curve for a higher year-end turn. I can think of two reasons:

  1. Many traders have probably never seen a high year-end turn priced across the curve in a non-stress environment. They could think this 2017 increase was a one-off. You have to go back over 15 years to see a turn further out the curve in a non-stress environment, after all.
  2. In this world of algos, it may take a little time. If you develop some model you think is awesome in this post-recession, new normal environment, I’m not sure you go back as far as 15 years to backtest it. So to an algo, this could just be a “six+ sigma event.” Umm… No.

It may take a little while for those two large groups of traders to catch up. I would think most normal traders, swap desks, algos, etc would have some “balancing mechanism” if they keep getting hit on their EDZ bids. “Balancing mechanism” = stops. I do think the markets will start coming around to the idea that the EDZs should be worth more.

The other minor consideration is that the year-end turn can be anywhere from 2 to 4 days. The market seems to ignore this further out the curve. But you should be aware that in some years, depending on the calendar configuration, the year-end turns can vary in length. I think this does come into play more in the immediate year-end turn. The 4 day year-end turn could be one of the reasons LIBOR has widened as much as it has so far this year.

It may be less likely we get a similar magnitude move next year. In any event, if I’m wrong, it will cost next to NOTHING! It’s not as if there was anything more than the minimum amount priced into the year-end turn throughout the curve. We just saw the LIBOR fixings the past three weeks increase 17 basis points! Even if you don’t want to do a specific year-end turn trade, you should consider checking your EDZ exposure up and down the curve, to make sure you are not overly long the EDZs. There is a tail risk on the turn getting priced back in – if you don’t want to profit from it, make sure you don’t lose money by not taking precautions.

The year-end turns still look like they are below 1bp on many parts of the curve. I think the turns should be at least 2bps, with upside. Here are some things you can consider (around whatever macro views you have):

  • Calendar spreads. If you like steepeners, buy calendar spreads that end with a Z contract. If you like flatteners, sell calendar spreads that start with a Z contract.
  • Single Flies. If you want to be long curvature, consider buying 3mo, 6mo, 9mo, or even 15mo or 18mo flies around the Z contracts. Similarly, if you want to be short curvature, consider selling flies that start or end with Z contracts.
  • Double Flies. You can buy 6mo double flies starting with an EDM, or sell 6mo double flies starting with an EDZ, depending on your curvature view. This assumes adequate margin and transactions costs.
  • ED vs FF. You can always sell EDZ8 vs FF in a reasonable structure. An example could be buying the EDM8-Z8 spread vs selling the FFN8-F9 spread. But you can also consider buying EDU8-Z8 vs selling FFV8-F9 or selling EDZ8-H9 vs buying FFF9-J9 (or K9).
  • Options. If you think that EDZ-FF could blow out over 10bps, this makes owning EDZ8 puts that much more attractive (with the intention of holding to expiry). The markets are pricing LIBOR-FF to fall about 7.5bps in Q1. So selling some put structure on say EDU8 or EDH9 to buy some EDZ8 put structure could make sense.

It’s not too late to get on some short EDZ exposure in attractive structures. We have thin markets this week, so you may be able to get some good turn looks on at favorable levels. There is a minimum amount of year-end turn priced later on the curve, so the risk/reward is attractive.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.