As we have seen the past few weeks, blindly smoothing the interest rate curve can have disastrous results:
- Meeting to Meeting: We have seen for some time that the quarterly meetings are going to be priced higher than non-quarterly meetings. When doing an analysis of a trade structure, it’s important not just to see how many meetings you are getting long or short, but which meetings you are getting long or short. As we’ve sold off, the absolute discrepancy between the quarterly and non-quarterly meetings have grown. At some point, it may be more attractive to take a relative overweighted non-quarterly meeting vs a quarterly meeting trade.
- Quarter to Quarter: We’ve seen recently the 3mo double flies have massive ranges. While the meetings in the next year or two are far enough away for there to be uncertainty, there are going to be certain meetings the market gravitates toward. When 2 hikes per year were expected, the markets gravitated towards June and Dec. So things like H7 and U7 3mo double flies got to extreme levels. Now that we’ve broken the “2 hikes a year” mentality, these structures have normalized, as the H and U meetings have become more likely.
- Semi-annual to Semi-annual: We’ve also seen where the 6mo flies (and double flies) curves can be very jagged. The Z contracts in particular seemed to be better offered, and the 6mo flies around the Z contracts were much higher than the 6mo flies around the M contracts. This could be in part pricing in a larger year-end turn.
In general, the closer to the front of the curve, the more the above discrepancies can increase. This is because the markets can have a stronger view on nearer term events. We could easily get some events that could cause a disruption in the curve, like Article 50, a political deadline, an economic data anomaly, central bank timelines, etc. As we get further out the curve, we would expect the strength of these potential disruptions to lessen as there is less clarity. However, because the year flies tend to be the most “smoothed,” we can sometimes seem effects in the front of the curve carry over to the back of the curve. We should look for these opportunities to fade. While nearer-term event risk is much higher, as we get further out things start looking murkier.
All other things being equal, I think the longer end of the curve is better for doing certain types of “smoothing” relative value trades. While the length of time to maturity and lack of clarity in the long end can provide good reasons for “smoothing,” there are also reasons for parts of the long end to get kinked. For example, currently the BOJ is targeting 10yr JGB yields. While they are not targeting the US curve, participants in the market are going to do 10yr Japan to 10yr US cross-currency trades. So we’ll see a depression in curvature at the long end of the curve. There are other reasons for the longer end depression as well: people trading convexity in the Eurodollars, a break between where Eurodollars and swaps are traditionally used to build the yield curve, and if aggressive hikes start becoming priced we could have the possibility of an inverted curve after an aggressive hiking cycle. We are not close to pricing in an inversion, but we should always keep this in mind as potential causes for additional kinking in the curve. Knowing is half the battle.
Fixed income investors are very savvy at picking up value, so you will see them gravitate towards parts of the yield curve that provide additional value, which results in a smoother curve. So for now, I think we are being provided with a good risk / reward opportunity to pick up some roll on the curve, and a cheap look at a shake-up in the curve pricing…