I only didn't include dividends in the measure because I was trying to show that the real earnings growth rate for the S&P (1.7% for my source), over 130 years, correlates well to the real S&P annual return of 2.17%(excluding dividends). My point was that earnings matter a lot and investors shouldn't just pay any multiple for them as most investors on here suggest. The real S&P price seems to grow at a similar rate to its real earnings over the very long term. So knowing that, wouldn't it be prudent to focus on earnings the most when trying to determine value and expected return? This is why I like the shiller PE so much. It's also why elevated corporate profit margins bother me so much also. Maybe I'm wrong. Does this makes sense to you?
I would expect in general over the medium term, price growth and earnings growth should be similar, but earnings is a 'soft' measure, so that probably explains at least some of the differences.
I understand using shiller/cape PE to smooth out results and look for bubbles, but I'm a bit wary of comparing it with long term values. Although, if you're trying to work out the earnings in one year, I think straight 100/PE would be more accurate than 100/CAPE.
Say if a company earned, over 10 years
$1 $2 $3 $4 $5 $6 $7 $8 $9 $10,
and its market price is 100, its CAPE is 100/((1+2+3+4+5+6+7+8+9+10)/10) =100/5.5 = 18.2. Its PE is 100/10=10, So if you know its market price is 100, and you want to know what its current year earnings are, its 100/10=10, if you do 100/18.2=5.5 (which is only the average of its recent earnings).
In general, earnings increase with inflation, and as the economy grows, so if you use CAPE, most of the time you will get an underestimate.
As to if they should buy at these levels or not, as far as I can deduce from what I've read, simply putting money as soon as you get it, into index funds, will beat the vast majority of professionals and amateurs who invest in the stock market.
Some people might be able to beat that, but the vast majority of people who try, won't. So its hard to say that following the above strategy is bad. Personally, I spend a lot of time studying individual companies, and I tend to buy companies with low PEs, high dividends, high ROE and high dividend growth - so even if I feel the market is expensive, I can usually find some good underdogs.
Estimating the total return, is useful when considering an undervalued market, I find this a bit pessimistic, but wouldn't you agree that you'd expect shares to grow by,
(earnings)+(growth)+(inflation), so
100/23.6+1.7+1.5= 7.4%.
http://www.gurufocus.com/shiller-PE.php claims that by CAPE the market is overvalued by 41.3%.
So if it took 6 years for cape to get back to normal, you'd about break even, if it takes longer than that, you'd be increasingly worse off, for each year you were out of the market. The market has been 'above average' for almost all of the last 23 years - you could be waiting a very long time.