See the Excel file here: http://www.econ.yale.edu/~shiller/data.htm
The stock market is still very highly priced compared to historical standards (excluding the past 15 or so years).
See also the graphs on: http://en.wikipedia.org/wiki/Robert_Shiller
From some cursory chart exploration on yahoo, it looks like large-caps might not follow the market moves as much. Mar 2009 was about 50% of the S&P 500's current level, but KO only differed by 30% and JNJ only varied by 10%. I'd say solid dividend companies are not necessarily overpriced.
And as far as the market being expensive in relation to historics, again you have to compare apples with apples. There are more people investing now than 50 years ago, and there are more big companies. Both of these facts will drive up the price of the mainstream companies' shares.
I'd rephrase that as something like:
If 16 is the average, then the current market isn't historically overpriced.
It could still be overpriced or underpriced. The historical average is just that. Current fundamentals aren't necessarily "average".
http://www.google.com/finance/stockscreener#c0=PE&min0=1...
(There were 3891 total companies when viewing the whole market.)
The distribution skews high though, so it's quite possible that the mean is about 20. OTOH, just what does a "mean" mean for P/E? Capitalization-weighted? Earnings-weighted?
If you look only at companies with a market cap > $1B, there are 1626 firms, and the median is again about 16. Same with $10B - this seems remarkably consistent.
The Schiller figure he references might include all publicly traded companies. If so, it might suggest that less well-known companies (outside the 500 most popular) are over-priced. But since the 500 most popular drive the market, you can't conclude a crash from that.
Most large-cap companies I have been looking at are trading at 15 year lows for the P/E, and at close to 15 year highs for dividend ratios. They still might be over-priced in an absolute sense, but we haven't seen prices this good in 15 years.