I find this discussion frustrating because I disagree with both parties on major points. Having said that, I think that Professor Seigel has the better argument (video below).
I’ve often criticised Professor Shiller’s attachment to the 10-year P/E Ratio, which he calls the Cyclically Adjusted P/E Ratio (or CAPE). The problem is that both stock prices and earnings are cyclical so it makes sense that their valuation metrics should be cyclical as well.
The other problem with looking so far back into the past is that the CAPE begins to tell you less about current valuations and more about the past earnings trend. Professor Seigel makes this point and he’s correct: the elevated CAPE highlights how poor earnings have been since 2001. I don’t know how that portends a poor-performing stock market.
The other issue is that many people don’t see what Professor Shiller is doing. He makes it clear that he’s not trying to predict market tops or bottoms. He’s trying to assess a very long-term market outlook. This is something he makes very clear yet he’s earned a reputation for making accurate market calls which he never has.
If you followed CAPE you would have missed out on much of the greatest buying opportunity of the last 70 years. Shiller was bullish during part of the run but he turned cautious very early. In fact, Shiller is routinely credited for being accurate when he’s really being consistently cautious.
Don’t get me wrong — I think Shiller’s goal of making long-term assessments of the market is important. But I’m far more concerned with the question of where the stock market (and individual stocks) should be priced right now.
Personally, I think the yield curve does a better job than the P/E Ratio does.
More: Megan McArdle takes Shiller’s side.