Home Forums Main Forum What happens to the curve during a hiking cycle?

  • Author
    Posts
  • Curve Advisor
    Keymaster
    Post count: 612
    #1352 |

    The last hiking cycle in the US was almost a decade ago. So there are a new crop of traders who have never seen what happens to the curve during a hiking cycle. At the end of the day, it all goes back to where you think Fed rates will be at a point in time. There are going to be other factors involved in determining the shape of the curve, like risk premiums and convexity. But typically, the primary driver is going to be the path of Fed rate moves.

    When the Fed models rates, the result will look something like this chart (from paper by William English from Nov 7, 2103). It basically shows various Fed funds paths if you use various unemployment thresholds (5%, 5.5%, 6.5%, 7%), for a given level of inflation (2.5%). The exact details aren’t important. What I wanted you to notice are the following:

    * The earlier you start hiking (higher unemployment thresholds), the slower the rate of increase in rates AND the terminal rate is lower.
    * The later you start hiking (lower unemployment thresholds), the faster the rate of increase in rates AND the terminal rate is higher.

    The reason for this is that monetary policy works with a lag. So the later you start, the faster you will have to raise rates (to combat overheating) and the higher you have to raise rates to. This has implications for the shape of the curve as hikes approach.

    [to be continued]

    Attachments:
    You must be logged in to view attached files.
  • Curve Advisor
    Keymaster
    Post count: 612

    When you think about the current shape of the curve, think of the market curve as being the probability weighted curve of all the possible Fed paths. You may be saying to yourself, “that makes no sense – if you take the probability weighted path of the lines in the chart from the previous path, you get that the long term rate is over 4%. The markets are pricing in nothing even remotely close to that! The long term rate currently priced is nowhere near 4%.” Yes. But I never said those sample curves generated from the Fed models were the only paths. In fact, there are a number of things you need to factor in:

    • There is a non-zero probability that the Fed NEVER raises rates. In fact this may even be the highest probability path!
    • There is a non-zero probability that the Fed may have to PAUSE after one or more rate hikes.
    • There is a non-zero probability that the Fed may have to EASE (either before or after a rate hike).
    • There is a non-zero probability that the long term FF rate is significantly below 4%, and could be as low as 2%.

    Most “models” are not going to cover all possibilities. I don’t expect you to write out very possible outcome. But you should get into the habit of thinking about all the “real” major possibilities.

  • Curve Advisor
    Keymaster
    Post count: 612

    Let’s take a step back and think about how the curve has moved over the past year. We went from an environment where:
    * tapering ended in October 2014,
    * FOMC changed from “considerable time” to “patient” in December 2014
    * FOMC removed “patient” in March 2015

    These are all moves that denote a progression from later to sooner hikes. If we look empirically at the constant maturity chart below, we can see that the shorter end of the curve (ED1 and ED5) has sold off, while the longer end of the curve (ED13 and ED17) has rallied. This aligns with the theory “The earlier you start hiking… the slower the rate of increase in rates AND the terminal rate is lower.” The theory assumes “all other things being equal.” The terminal rates being lower is not always true – if we had gotten some spectacular growth, it’s possible the long end could have sold off more as hikes were being priced in. Over the past year, what we actually got was QE in Europe, weakness in Asia, and lower commodity prices. So the long end may have rallied more than “usual,” and the selloff in the front of the curve has been more muted than “usual.”

    [next: changes in slope and curvature]

    Attachments:
    You must be logged in to view attached files.
  • Curve Advisor
    Keymaster
    Post count: 612

    I want to discuss what happened to the slope the past year. As more hikes move from “later” to “sooner”, you would expect the year spreads in the back of the curve to decline, and the spreads in the front of the curve to increase. You can see the declining year spreads most in the ED9-13 spread. Further out the curve, you see a small decrease, but those year spreads were not as steep, so there was less to fall. What is interesting is that ED5-9 spread has also fallen somewhat dramatically. The Fed’s projections show the majority of rate hikes in 2017 (122bps in 2017 vs 100bps in 2016). However, the markets seem to be discounting a lot of the hikes in 2017. Some of the factors mentioned in the previous post that have led to the long end rally also affect the rest of the curve.

    You can see the first year spread (ED1-5) has increased over time. As we’ve seen with the single contracts, the move has been more muted than “usual.” It’s a little strange to have the Fed hike presumably starting this year, and there being only 3 hikes per year priced in at teh front of the curve. But each hiking cycle is different, so you always need to consider what is each time period covered by an ED spread carefully.

    Attachments:
    You must be logged in to view attached files.
  • Curve Advisor
    Keymaster
    Post count: 612

    Finally, I want to discuss what happened to curvature the past year. As more hikes move from “later” to “sooner”, you would expect the butterflies in the back of the curve to decline (as the slope declines), and the flies in the front of the curve to increase. You can see the declining year flies most in the ED9-13-17 fly. Strangely ED 17-21-25 fly has increased, which is a little strange, and the basis of some CA trades this year. But the most notable thing on the curve is the sharp increase in the ED1-5-9 year fly. Note that the fly has moved much more than the single contract or the slope. Further out the curve, it is typically the case that butterflies move the least, but near the front of the curve, butterflies can move more than the individual contracts.

    All three of {direction, slope and curvature} tell a similar story, but from a slightly different perspective. When you look back at all three charts, it’s hard to glean much from first glance at the chart of just the contract prices. But as you take higher derivatives of the first chart, you get alternate views. And sometimes a slightly different view may give you a whole new perspective on the markets and possible trading opportunities.

    [next: what to expect going forward]

    Attachments:
    You must be logged in to view attached files.
  • Curve Advisor
    Keymaster
    Post count: 612

    What can we expect as the hikes get closer? Some of the trends we’ve seen above will continue:
    * Whites (and reds) will continue to sell off
    * The front year spreads will continue to increase and the year spreads further out the curve will continue to flatten
    * Curvature around the reds will continue to increase and curvature further out the curve will continue to decrease.
    As hikes start looking more distant, the opposite will occur.

    The one thing I would caution against is in paying TOO much attention to historicals from previous hiking cycles. There may be some algos that are historical-based, and there might be a number of old-timers in the market who will also be biased towards their previous experiences. However, just as this recovery has been “unusual,” so too may the path of Fed rate hikes and the changes in the shape of the curve during the hikes in the current hiking cycle.

    When thinking about the shape of the curve in a hiking cycle, you need to focus on the following:
    * How strong is the market’s projected path of growth?
    * How strong is the market’s projected path of inflation?
    * What is the market’s assessment of the Fed’s reaction function to the above?

    [to be continued]

You must be logged in to reply to this topic.