While diversification gives you a less volatile portfolio, it also provides you with a smaller return than 100% stocks (although some would argue this statement).
I wont argue, but I'll disagree...
Several of the "asset allocation" guys point out that high diversification portfolios can enhance returns while concurrently reducing volatility.
But it clearly depends on what you're using for diversification classes, how much, and what periods you measure.
The three down years we just had were pretty good for bonds, so a 50/50 or 60/40 stock/bond mix would have done way better during that period than a pure stock portfolio. While reducing overall volatility.
REITS, emerging markets and small caps have done far better than the s&p500 over the last couple of years. Incorporating some of those would have improved your returns, again while reducing volatility.
Riskier and less risky classes besides US stocks CAN improve your returns while reducing portfolio volatility, providing you balance correctly, rebalance periodically (every 1-3 years), and choose classes that are not highly correlated with one another.
Here are some asset classes and correlation influences I incorporate into my asset allocation spreadsheet; this information comes from Gillette Edmunds book. Not as thorough as Bernsteins, but a one sitting read without as much brain burn:
- Emerging Markets, US Stocks and foreign on different "tracks".
- Small US and Large US stocks are on the same one.
- US and Foreign stocks on different ones.
- Emerging market and foreign are somewhat correlated.
- Real estate and oil/gas on similar ones.
- Bonds and stocks in the same country often track similar
- High inflation hurts stocks and bonds in the short term, then can improve stocks by higher product selling prices and bonds by accompanying higher interest rates.
- Declining inflation helps stocks and bonds in the short to medium term.
- Rising currency value helps both stocks and bonds in that country.
- Declining currency hurts both stocks and bonds in that country.
- Money markets, tbills and cash track similar to bonds.
From this, a smidgeon of foreign stocks (which can be further sub-split to small and large cap), some REIT's and/or oil and gas pumpers, a helping of domestic bonds, and perhaps a little foreign bonds and/or emerging market stocks. Then a precious metals cherry on top?
Its also worth it to note that with our current low inflation and rates, its likely both may rise and fall over our retirement period, inflicting benefit and damage on the US markets; these may have inverse relationships on foreign issues as those become more expensive or cheaper relative to US investment instruments.
Plus our low US dollar value right now will (again hopefully) reverse itself, sending a few ripples.
With these in mind, owning any long term bonds or a very heavy intermediate term bond (10+ years) may be a bad idea. Further, Bernstein showed that the risk/return profile of 10+ year bonds isnt very appealing most of the time.
Several "portfolio's for every season" have been posted here, but:
10% to foreign issues
3-5% emerging markets
5-15% reits
2-3% precious metals (perhaps)
3-5% high quality foreign bonds (perhaps)
10-20% small cap value
remainder of portfolio split 60/40 between s&p 500 or total stock market and short term corporate and/or TIPS
You may wish to use large cap value instead of s&p500 or TSM for your major US stock holding. Higher returns and dividends over long time periods.
Rebalance between winners and losers every couple of years. Rinse and repeat. Some will argue that rebalancing all the time or monthly/quarterly gets you something. Others note that some assets have performed great for many years straight and awful for many years, so frequently rebalancing to/from those would hurt. I think every year or two might do the trick.