- Most people look at Shiller's CAPE to value the market.
- Others look at the POPE (Plain Old PE).
- Both may be missing the peak.
- Here's something with a much better track record at calling peaks.
Is the stock market close to its next peak? After 30-odd new S&P records this year alone, you may be wondering that, too.
There's no shortage of bears calling a top now. Prof. Robert Shiller has been widely quoted as saying that the market is "pricey" at the moment.
How can he tell? You use a stock's P/E ratio to judge if that stock is pricey or not. You can use the market's P/E in the same way. That's what Prof. Shiller does, only he uses a custom measure he calls the CAPE (Cyclically Adjusted PE).
THE CAPE AND THE POPE AS WARNINGS
What's the difference between the CAPE and the POPE (Plain Old PE)? Compare them side by side:
How has the CAPE done in the past as a warning?
The four vertical dotted lines show when the stock market turned down the previous four times. You can see the CAPE, while moving nicely with the S&P, didn't warn us ahead of time of the impending drop.
Nice, in an academic kind of way, then, but not useful to investors.
Did the POPE do any better as a warning tool?
Not really. You could make the case that it gave a warning in 1999, but one out of four does not a good warning make.
There's another problem: what do you use as the average to figure out if the current PE is "too high?" The average PE has been significantly influenced by changing interest rates:
The dotted grey arrows show the PE went down in the days of rising interest rates, then it went up as the Fed dropped those rates.
So... here you are in 2014, with record-low interest rates. What's "normal?"
And how close is today's PE to normal? Look at where the orange line ends. Is that "high?" Sure doesn't look so.
Prof. Shiller contends the CAPE calculation smooths out those wild swings and gives us a better view. Does it?
Look at the teal line in the first chart. Now, ask: what's "normal?" The only time that line had any degree of flatness to it is between 2004 and 2008, when it spent an extended period above 25. Is that "normal?"
Prof. Shiller says the market's "pricey" because the CAPE is above 20. But it's been above 20 all the time since the new millennium (except for a brief period in the Great Recession).
How does that help you as an investor? If "CAPE over 20" is the alarm bell, shouldn't the market have been crashing for 14 years running? We know that hasn't happened.
So, pinning down an average as a benchmark, to determine whether the market is overvalued or not, is next to impossible.
Therefore, neither PE measure is good for calling a peak. (Unless you simply call a peak for 120 months, hoping you'll eventually hit it and people will forget the 119 times you were wrong.)
IS THERE SOMETHING BETTER?
There may be: earnings.
Using the same Shiller/Yale database, let's plot the S&P 500 against its TTM (trailing twelve months) earnings:
Look at that: every time, the dotted green ovals show there's some sort of warning before the market tanks.
Granted, it's not perfect -- the 1998 earnings drop was way too soon, and the 2007 drop didn't give us all that much warning.
But... peaking earnings did significantly better at predicting a market peak than either of the two PE measures.
Look at Prof. Shiller's inflation-adjusted earnings, both actual and smoothed for his CAPE calculation with a 10-year moving average:
Three things stand out:
- Long term, the earnings of the 500 companies in the S&P 500 are rising.
- However, the wave pattern suggests that in each cycle, earnings climb to a point of overreaching, at which point they collapse.
- Each peak is higher than the one before in this era of low interest rates.
It's as if the collective earnings engine of the economy (of which the largest 500 corporations make up a big part) keeps growing, but not evenly. Each cycle, it takes a deeper gulp of air before exhaling again.
So, the question you ask yourself at this point is: how close are earnings now to another peak? They're higher than the 2007 peak, so that's like a yellow light.
The rising peaks would seem to suggest this bull still has a little more headroom before earnings turn south and trip the full-scale red lights and alarm bells.
It's not perfect; there's still an element of fuzziness in the picture. (It's the future we're talking about, and nobody knows the future.)
But staring at the P/E charts doesn't give you this kind of clarity.
Bottom line: Peaking earnings of the S&P 500 index could give you a much better warning of the next market peak than trying to figure out how to compare its PE with some moving benchmark.