What Would It Take to Trigger a Secular Reversal in Bond Yields?
In the fall of 1981, the twenty-year US bond yield peaked slightly above 15% and has been zig zagging down through each successive business cycle since. During the last one hundred and sixty-years or so, the average secular, (very long-term) trend in rates has lasted around twenty seven-years. After thirty five-years of declining rates, the current secular bear is getting long in the tooth. Thus far our late 2016 forecast A Turn in the Tide: The Case for Rising Interest Rates, appears prescient as the low watermark for yields remains in place. The question naturally arises, as to what technical signals you should follow to reliably confirm the reversal.
Previous secular trend reversals
At the outset, it’s important to bear in mind that there have only been four previous secular trend reversals. This greatly limits the data points on which to base our analysis. The length of such trends also means that indications of a reversal are usually given several years after the final turning point. Nevertheless, it is possible to list four technical characteristics that have been present at previous turning points.
They are, price patterns, moving average crossovers, trendline violations and momentum characteristics. Let’s apply them all to recent action for the twenty-year yield and see where we come out.
Technical evidence required for a secular reversal
Chart 1, which features the trend of government bond yields back to the 1870’s, shows that there have been five secular trends since then. Three were down and two up. The average duration of two bulls was thirty-years and that for the two completed bears, twenty-five. Based on duration alone, that makes the most recent thirty five-year drop since 1981, ripe for a turn.
Using the benefit of hindsight, it is quite easy to identify previous peaks and troughs, but in real time we do not have that luxury. The trend of rates though, rarely reverses on a dime, instead slowing its upward or downward trajectory ahead of time as it adjusts to structural and institutional changes in the economy. Oil tankers take a long time to turn around, so do secular trends in financial markets. These transition periods take the form of multi-year trading ranges, the average duration of which has approximated twelve-years. Depending on whether 2011 or 2014 is taken as the reference point for the beginning of the current trading range (see the small horizontal lines), that would translate to a breakout in the next six-nine- years. Alternatively, we could consider the six-year average time between the secular turning point and the subsequent breakout. That would translate to a 2022 signal based on the 2016 low. Normally, secular reversals develop more or less simultaneously for the vast majority of maturities, though the data on which this conclusion is based is pretty scant. In the current situation though, many shorter-term maturities of five-years or less, bottomed in the 2011-12 period. That suggests that the base building process for the current cycle may be far more advanced than an analysis of the twenty-year series alone would have us believe. Bottom line, we should expect to see a breakout at any time.
Chart 1 Government 20-Year Yield 1876—2017
Crossovers of the 96-month Exponential Moving Average (EMA) have offered reliable signals of secular reversals.