Stocks are Really Hanging in There
Rising Bond Yields
With the 10-year US Treasury Yield approaching 4.3% today, a 15 year high, I thought it would be interesting to take a look at the historical relationship between this yield, mortgage rates, and the Shiller PE also known as the Cyclically Adjusted Price to Earnings (CAPE) and the total return CAPE (TR CAPE). A quick summary of these metrics and an explanation for TR CAPE is that the CAPE may be affected by changes in corporate payout policy and the TR CAPE corrects for this. More detail is on Robert Shiller’s website for those interested. But instead of showing these metrics, I will show the inverse of these metrics. This is for good reason as Antti Ilmanen explains in this interview:
if we take the discount rate for equities, like a simple way of thinking of the equity markets--it would be to look at the Shiller earnings yield. So this is the inverse of the CAPE ratio so one can think of that as the discount rate for the S&P 500 and that was basically ten percent forty years ago and then it came down to three percent, and that gave very nice returns as the repricing happened.
This is depicted in the chart below and in fact, this decrease in CAPE yield is even more profound for TR CAPE as it peaked around 15% and is currently 3.24%.
Why CAPE and not P/E
So why TR CAPE? In their 2007 paper “The Many Colours of CAPE,” authors Robert Shiller and Farouk Jivraj present this table showing the historical real 10-year S&P 500 returns from various starting CAPE ratios. Note that in all but the lowest CAPE decile, the average annualized returns walks down quite consistently between deciles. Something to consider with today’s TR CAPE at 33.25.
The authors further presented T-values for CAPE versus other popular valuation metrics and showed that across forecasting horizons, CAPE was the metric with the most consistent statistical significance.
Relationship Between 1/CAPE, Bond Yields and Mortgage Rates
The chart above juxtaposes both CAPE yields, the 10-year UST yield and mortgage rates over time. The reason for this layout is to help convey the intuition that historically these metrics largely walked in lock step at horizons of 5 years and more. I mention this because this presentation, with recessions noted, shows that during recessions the CAPE yields tend to go up and bond yields tend to go down. But these relationships are over periods of much less than 5 years. But when you look at longer horizons, you see that bond yields, mortgage rates, and both CAPE yields generally walked up between 1962 and 1982, which was the inflection point. Since then and until the beginning of 2021, all of these metrics generally moved downward with the notable exception being the dot com bubble which saw both CAPE yields bottom at 2% then moved to 4%, blew out to almost 8% during the GFC, and then settled in at 4% or lower for almost the last decade.
Inflection Point
The curious part about all of this is it appears that the long downward march in bond yields and mortgage rates came to an end at the beginning of 2021. This was followed by a slight uptick in CAPE yields but they have moved back down to 3.01% and 3.24% where they are today. Yet this slight upward movement in these yields pale in comparison to the more than doubling in both mortgage rates and 10 year bond yields. Moreover, the historical precedent is that CAPE yields generally follow bond and mortgage rates upward, as they did in the 15 years prior to peak bond yields in 1982.
Lessons to be Learned
What can we learn from this relationship? We may want to take everything we thought we knew about the historical relationships between asset prices and inflation that used to keep us warm at night, thinking we understood how the investing landscape works…
And kick it off a bridge!