The thing about 1997-2000 is the implied yield of the S&P by either dividends or CAPE10 was 2-4%. At the same time, expected real bond returns were 3-5% or higher. You could make a good case then that the 30 year 4% TIPS bond would outperform the S&P with a CAPE of 40.
Now, bonds are set for a real return of 0-1%, while US stocks should be looking at 3-4%. It's hard to make a case now that things are overvalued, instead of just higher than they used to be. You can say that ex-US should do better going forward (I am investing as if it is a good chance) but that is not certain. The problem with saying something is "overvalued" is that is a relative term, and you must then define what will be better. It doesn't seem obvious to me what the better choice is.