A few points worth mentioning here:
1. Basing an investment strategy on a recent 15-year period of returns is probably a bad idea. Else, we should all invest entirely in long-term government bonds and small-cap value stocks.
2. Looking at annualized returns of each component is not particularly helpful, as we should be more interested in how adding a new asset class (in this case, foreign stocks) affects the overall risk and return characteristics of the entire portfolio. In this simple 2-asset class universe, we have the following annualized returns and standard deviations in annual returns for the 15-year period of 1999-2013:
By going from 100% US stocks to 50% US stocks, you would have lost ~0.06% in annualized returns, or ~0.91% over the entire 15-year period. There was a similarly-negligible increase in portfolio volatility. I mean, let's call it a wash. Further, approximately ALL of the performance differential between 100% US and 50% US can be explained by what used to be a much larger difference in expense ratio between the two funds. BTW, as of now, the ETF variants of these funds, VTI and VXUS have an expense ratio differential of 0.09% per year, at 0.05% and 0.14%, respectively.
I believe that Mr. Bogle is wrong in advocating such a heavy US bias, however I acknowledge that it probably doesn't matter very much compared to other much more important decisions in developing a suitable asset allocation.
- DDB
"We have to encourage a return to traditional moral values. Most importantly, we have to promote general social concern, and less materialism in young people." - PB