According to my latest calculations, the S&P 500 has at least 143% upside to 4,000 in order to reach the historical average ratio of earnings yields vs the 10 year treasury. It could go up closer to 300% to 6,700.
I do not actually believe this.
Nor should you. The main takeaway is that this and ALL OTHER such “historical” valuation projections are mostly bunk. This seems particularly clear today, when the underlying data mostly just shows how uncharted our current territory is.
So I was messing around with S&P 500 earnings data back to 1960. The most interesting chart was the S&P500 annual earnings yield (basically the “interest rate” earned on a share of the S&P) versus 10 year treasury interest rates. It is crazy-wacko-sky-freakin-high! It hasn’t been this high since the 1974 meltdown and the period thereafter.
Put in plainer terms, the relative attractiveness of the S&P500 versus a “safe” investment like bonds is high and has been getting higher as of late. Which does suggest the market is at least not in bubble territory from a relative valuation perspective.
Versus 1974, the earnings yield (6.47%) and treasury yield (2.66%) today are a both a whole lot lower – the 1974 earnings yield was 13.64% vs treasury rates of 7.5%. The main difference is inflation. It was running wild back then. It is running perilously close to deflation today. But you are getting a whole lot more protection from inflation in the S&P500 than you’re gonna get from bonds at these levels.
The other thing that jumps out of the earnings history is a huge discontinuity around 1995.
The chart below is for REAL S&P500 earnings (inflation adjusted to 2005 dollars). Real earnings just exploded after basically doing nothing from 1960 on. I am still mulling over the “why’s of that (tax code changes? Monica Lewinsky?) but it is very clear that historical comparisons back to earlier periods are apples vs. oranges.
So what about that 4,000, 143% upside target for the S&P 500? That is how much the S&P500 would have to go up to take the earnings yield to match the current 10 year treasury yield. Actually that is being conservative. A look at the first chart shows that the earnings yield was BELOW the treasury yield from 1980 to 2000. So arguably the S&P500 has to go up nearly 300% to @6,700 to get to the -1% levels seen over that 20 year period. Whooee! If I can just stretch that argument out into a plausible sounding book, I’d probably be set as the resident Pollyanna on the speaker’s circuit (at least until the next major bust).
Actually I am mostly taking all of the above as offering some comfort at current valuation levels. I am still trying to work out (for my own sake) whether we really have much upside from here. One concern is that the market didn’t deliver decent returns until well into the 1980’s after the 1974 bust/recovery. But then again, I just spent a lot of time above questioning the validity of that sort of historical pattern-matching. And the inflation picture today is totally different. Or is it?
I got jammed up against my publishing deadline today and wanted to get these charts out before earnings devour us all. So that all for now. I will circle back on it later on, however.