copyright1997reloaded
Thinks s/he gets paid by the post
On Feb 1, CAPE10 had reached 34.1
By Fed 9 it had only dropped to 31.7 - still in nosebleed territory exceeding any other time except mid 1997 - mid 2001. This is still above the Nov 1 2017 value of 31.2.
Far be it from me to think that I know more than Shiller. But (isn't there always a but) the Cape 10 measurement has some interesting characteristics as measured:
1. Graham & Dodd in their 1936 Security Analysis work: https://www.amazon.com/Security-Analysis-Foreword-Buffett-Editions/dp/0071592539 originated the idea of eliminating economic cycle swings in analyzing PE ratios and suggested looking back over a longer period of time. But (there it is) they weren't saying 10 years was the only period, but also suggested smaller time frame measurements (e.g. 5 or 7 years). There is nothing magical about a 10 year look back period.
2. CAPE10 uses GAAP earnings, which are generally lower than operating earnings and also which are more severely impacted by the 2008-2012 "great recession". (I would also guess that they will impact earnings now as companies take one time tax hits to deal with non-repatriated funds).
3. There have been other accounting changes which makes the comparison of earnings today different than long ago. For instance, FAS 142 changed the way goodwill is accounted for.
As you mention above, the CAPE 10 value topped 30 during 1997-2001. It first topped 30 in June of 1997, and topped 36 in March of 1998...a full two years before the market topped.
Having said all of that, your point is valid and a high CAPE 10 value is certainly a warning signal that we should all be aware of.