The investing mantra these days is to keep a well balanced portfolio. Before reading this book I understood this approach as a mechanism for reducing risk, but have never seen a thorough explanation of the thinking behind it. Mr. Bernstein provides a step-by-step analysis of how portfolios perform in the long run (which is the key to his strategy). Here are the key points:

• Each asset class has an expected return rate, which determines how well that asset will perform in the long-run.
• Each asset class has a risk associated with it, which is defined as the standard deviation of the yearly return rates of the asset. The risk determines how long you may have to wait before the asset performes up to its expected return rate.
• The performance of two asset classes can be compared to determine whether they tend to go up and down together, or behave independently. This can be expressed as a number, known as the correlation of the two asset classes.

Then he shows how if the performance of two assets is uncorrelated, by holding the correct proportions of those two assets (say stocks and bonds), you can actually achieve a return rate higher than either of the single assets, and with a lower amount of risk. Expanding the concept to multiple assets, you can choose how much risk you are willing to take and (with the help of a computer program) select a portfolio that should provide optimum returns for that amount of risk.

The key to this improved performance is the annual rebalancing of the assets. You take money out of an asset that has been doing well, thus protecting it from the inevitable reversal. You add money to the asset that has been doing less well, anticipating an upturn. If the assets behave oppositely from each other, you can see how this strategy could lead to improved performance over the long-run.

If your strategy is to invest your money in a tax-free account, and only look at it once a year, I have no doubt that this approach is a very sound one. However, the validity of this strategy as being the best one is based on the assumption that stock picking is impossible, and that the market can't be predicted. He admits in one section that if one asset class (say small-cap stocks) is doing better than others, that it's often better to leave the money there rather than rebalancing it, because momentum effects do seem to exist. And for those of us who remember the prolonged down market of 2000 - 2002, it's hard to argue too convincingly that leaving your money in the broad stock indexes was a sound strategy.

If you believe in this investing approach, then this is an excellent book. But if you believe (or want to believe) that there is money to be made in "playing" the market, then you'll want to augment this approach with other strategies.