Curve AdvisorKeymasterJanuary 7, 2015 at 8:23 pmPost count: 612
One of the things I always say is that when you look at historicals, make sure your historical comparison period is relevant. This comment applies to looking at recent historicals (i.e. 1 year). However, what I wanted to discuss today was looking at far back historicals. In this case, looking back at the last tightening cycle in 2004. There are some similarities (aggressive easing to the extremely low levels of FF rates, potentially large amount of hikes, etc). However, there are some key differences:
* the many think the longer term FF rate is much lower than the FOMC forecasts of 3.75%.
* the pace of hikes will probably be very slow. In 2004, it was not unreasonable to price in odds of a 50bp move.
* the FOMC is not looking to hike at a measured pace – in fact, it is possible they may intentionally go out of their way to be unmeasured.
* we are now on the zero bound. Some claim this makes the Fed less likely to hike, while others claim this makes the Fed more likely to hike.
* the economic situation is different.
* the personalities are different.
* the global backdrop is different.
The first three things are the most important – as once the Fed starts hiking, these factors will determine the shape of the hiking curve. When the individual factors are different in the two time periods, you can get very different curve shapes.
[to be continued]
Curve AdvisorKeymasterJanuary 13, 2015 at 2:08 pmPost count: 612
Below is what I wrote in this past weekend’s CA:
DIFFERENCE IN HIKING CYCLES. I started getting the “Danger, Will Robinson!”signals from the market last month. In particular, this hiking cycle, while it may seem very similar to the 2004 hiking cycle (aggressive easing followed by 300+bps to get to “normal”), is actually very different from the 2004 cycle – in particular, personalities, triggers for liftoff, global backdrop, strange structural/cyclical patterns, increased potential for policy error due to conflicting signals, potential pace of hikes, and terminal level of rates. The last two are particularly important for determining the shape of the curve. The really tall fly curves we were used to in a hiking cycle is no more. When the slope is going to be capped at 25bps a quarter (or 30bps if the markets get really “crazy”), there’s only so much room for a fly to go particularly high. The previous hiking cycle was “measured” and the FOMC now seems to think this is bad. An unmeasured hiking cycle also makes for more uneven fly trading, and should result in potentially lower flies in the reds. There also is a question as to whether the terminal funds rate is closer to 1.75% or 3.75%. It’s very possible we could get bimodal fly curves (and possibly trimodal) going forward. And the increased possibility of a policy error in the face of a poor global outlook means that eases can’t be ruled out 1 year or so after a hike.
Curve AdvisorKeymasterApril 6, 2015 at 10:22 pmPost count: 612
If you are going to look at historicals, one of the things you MUST do is to mark the important dates on the calendar. For example, from the past 7 months:
26-Sep-14 Bill Gross resigns
15-Oct-14 Equity & Ebola rally
29-Oct-14 Tapering Ends
17-Dec-14 FOMC changed to “patient”
18-Mar-15 FOMC removed “patient”
Some of the dates are just important events that may have only affected the curve temporarily (like the first two). But others could be the start of a “regime changes” – a more substantial change that could potentially alter the way the curve trades (the last three items). When you look at historicals, you can “blindly” just look at things like a “1 year” chart – barring any evidence, this is a reasonable thing to do. However, when you look at the charts, you should think about what the key dates are, and note if the trading has changed since those key dates.
Curve AdvisorKeymasterJanuary 7, 2016 at 2:36 pmPost count: 612
When looking at a historical range on an ED structure, you can’t just look at the range and think that is static. There are changes that go on all around the curve. So you need to factor in what is occurring in the rest of the curve, within the context of the historical range you are looking at. Here are some examples of adjustments you can consider:
* If you have on a fly trade in the middle of a historical range, but the curve is at new lows in terms of flatness, you should adjust down your entry and exit levels. The height of a fly is a function of slope. It is less likely for the fly to get back to its highs, when the fly is already high relative to the current slope.
* Always keep in mind that the contracts roll down the curve. If you hold the ED10-14 spread for three months, the relevant historical comparison period would be the historicals for the ED9-13 spread (and not the previous ED10-14 spread). As time passes, think about how the old historical range will gradually change into the new historical range.
You must be logged in to reply to this topic.