The importance of asset allocation - how a portfolio is divided among various asset classes - is widely understood. Experienced investors know that asset allocation is the key factor in portfolio performance and that superior risk-adjusted returns are achieved by combining uncorrelated asset classes in a balanced portfolio. Disciplined, methodical asset allocation has proven effective through all market environments. There is little disagreement about any of these points. What does arouse controversy is the question of whether asset allocation ought to be a passive or active activity.
Proponents of passive allocation argue that markets are efficient and that it is impossible to predict asset class performance. Therefore, it is unwise to engage in active asset allocation, especially in light of the higher transaction costs and tax liabilities that result from more frequent portfolio changes. Passive asset allocators believe that investors should determine a suitable mix of asset classes, rebalance periodically, and spend time on other pursuits. To a practitioner of this style of investment, active asset allocation seems to be just another name for market timing.
There is a great deal of validity to a passive asset allocation strategy, especially compared with the flawed approaches so many investors continue to pursue. However, a strictly passive approach to asset allocation fails to account for the fact that asset classes periodically reach extreme levels of overvaluation and undervaluation. These situations don't occur often, but when they do occur, they provide opportunities for an alert investor to make tactical asset allocation adjustments that have a high-probability of increasing returns or reducing risk.
Extreme Asset Class Valuations over the Past Five Years
The past five years have provided several examples of extreme asset class valuations - the most glaring of course being the overvaluation of large cap growth stocks in general and technology stocks in particular in the late 1990s. At the end of 1999, the price-to-book value multiple of the S&P 500/Barra Growth Index reached the astounding level of 14.7x (versus a long-range average of approximately 3x), creating one of the most unfavorable risk/reward relationships in financial history. Fortunately, value-conscious investors had many attractive alternatives in 1999, including bonds, real estate investment trusts (REITs), value stocks, small-cap stocks, and even cash, which at the time was yielding over 5%.
More recent examples of extreme asset class valuations include:
- The undervaluation of REITs in late 1999/early 2000
- The subsequent overvaluation of REITs from mid-2003 to early 2004
- The undervaluation of emerging markets stocks beginning in late 2001
- The undervaluation of high-yield bonds throughout 2001 and 2002
- The overvaluation of investment-grade bonds beginning in mid-2003
Each of these instances of overvaluation or undervaluation was recognizable using relative valuation analysis at the asset class level. Relative valuation analysis compares the current valuation of an asset class to both i) the historical valuation range for that asset class and ii) the current valuations of other asset classes.