Anatomy of a Bottom

"no Mr Bond, I expect you to die"



The bond market did talk, and recently experienced a steep sell-off, one that had been forecast here by Macro Man, who charted the long bond quite frequently from late Summer onwards, and predicted some pain ahead for US fixed income investors. Eventually this came to pass, although the trigger for the sell-off was an unexpected one to some. Perhaps the removal of US election ambiguity was enough to persuade many to sell, or perhaps there were real macroeconomic arguments that will be proven sound in time. What is clear, however, is that the selling seems to be over for the time being. We say this despite the crescendo in the media (mainly from sell side analysts) that “interest rates are going higher”, and that there is a “great rotation” out of bonds and into equities.

Why are we so confident that a bottom is in for US fixed income? Price is News, my friends. Let’s take a look at some charts, and we will review several different bottoms, for chart watchers who enjoy such things. It is often easier to identify Bottoms than Tops, perhaps because of a variety of useful technical signals, such as RSI < 20, declining volume indicating seller exhaustion, and of course MACDs.

After poring over many signals, we believe that the bottom is in for the following instruments: the closed end high yield muni fund IQI, for example, put in a low on the morning of November 14th, after two very high volume selling days (11/10 and 11/14). We were buyers 11/14 and 11/18.

Intrepid buyers on 11/14 have been rewarded with a 3% rally to go along with the 6% yield. It is interesting that RSI never < 20, but this is the muni market, it’s mainly the province of professional fixed income investors and not manic-depressive (or 12 year-old) hedge fund traders. Other ETFs that are more widely traded took longer to form convincing lows. For example, the muni ETF MUB and the high grade corporate bond ETF LQD both made new lows in December, and this happened long after the RSI had reached its nadir <20. The RSI made a double bottom in MUB, after which the ETF made a V-shaped bounce on high volume, a clear sign that the bottom is in:

LQD carved out a long double bottom formation after RSI dipped below 20 on 11/14; high volume down days in early December and mid-December specified a double bottom from which a recovery is now in progress:

We were buyers of LQD on high volume selling days, 11/18, 12/1 and 12/14.

Finally, the Long Bond, a subject of speculative activity via the futures and options market, took the longest time to recover. High volume selling days in mid November and early December eventually led to the formation of a long bathtub-shaped bottom in the second week of December. We began buying TLT on 11/14, and continued on 11/23, 12/1 and 12/9.

We note our buying dates to illustrate four things: 1) RSI <20 alone is not always a good guide in identifying a bottom, especially under extreme selling pressure, but it will usually be coming soon. 2) Beginning to buy and continuing to buy in tranches when RSI is <20 usually does work out OK a month or two later. 3) Waiting for RSI to reach 50 again will result in wasted opportunities. 4) Instruments that should “trade together” b/c “they are all bonds” in response to a stimulus such as “perception of rates going up” will not always do so in lock step, perhaps because the investor mix in IQI and MUB is different from that involved in LQD and TLT.

So what happens next to Mr Bond? The charts alone would seem to be telling us that a bottom is in and a slow recovery will occur. Other factors in play in the short-term are pension fund flows (already trickling into the market and putting the charts into a much rosier light this week) and COT positioning (commercials are long the 10 year, while non-commercial speculators are massively short as Abee illustrated yesterday), which sets up the possibility of a powerful short squeeze in the next week or two.

In support of the idea that bonds will continue to rally on mean reversion is the fact that long bond volatility of the magnitude seen during the recent panic was observed only in 2008 (GFC), 2009 (QE1), and 2015(taper tantrum), all very extreme events.

The longer-term picture is more difficult to predict, and will be the topic of another post here in January. Political factors and the Fed come into our thoughts when we look out into the future, so perhaps we will save the second post “Mr Bond Decides to Die Another Day” for the week of the Inauguration. Nevertheless, as a teaser, take a look at the gaps in the chart for 10-year yield, around 2.20% and 1.90%. Gaps get filled, my friends…

This post was written by Leftback