In a 1986 article, a group of researchers raised the shackles of many an active portfolio manager by estimating that as much as 93.6% of the variation in quarterly performance at professionally managed portfolios could be explained by the mix of stocks, bonds and cash at these portfolios. While this statistic is open to question, there can be no denying the importance of the asset allocation decision to overall portfolio returns. While the researchers looked at the allocation across financial assets alone, we would define the asset allocation decision much more broadly to include real assets, including real estate, and in the most general case, human capital.

The asset allocation decision follows logically from the client assessment in the previous section. Having understood the risk preferences, cash needs and tax status of the investor, the portfolio manager has to decide on the mix of assets that maximizes the after-tax returns subject to the risk and cash flow constraints. This is what we would term the passive approach to asset allocation, where the investor’s characteristics determine the right mix for the portfolio. In coming up with the mix, we draw on the lessons of diversification; asset classes tend to be influenced differently by macro economic events such a recessions or changes in inflation, and do not move in tandem. This, in turn, implies that diversifying across asset classes will yield better trade offs between risk and return than investing in any one risk class. The same can be said about expanding portfolios to include both domestic and foreign assets.

There is, however, an active component to asset allocation, which leads portfolio managers to deviate from the passive mix defined above, and that is market timing. To the extent that portfolio managers believe that they can time markets, i.e., determine which markets are likely to go up more than expected and which less than expected, they will alter the passive mixes accordingly. Thus, a portfolio manager who believes that the stock market is over valued and is ripe for a correction, while real estate is under valued, may reduce the proportion of the portfolio that is allocated to equities and increase the proportion allocated to real estate. In spite of their protestations to the contrary, most portfolio managers engage in some market timing, and the high profile of market strategists at all of the major investment firms suggests that the asset allocation decision is an important one. The reason is very simple; a successful market timer earns tremendous returns.

Its appeal to investors, notwithstanding, market timing remains an elusive dream for most. Looking back at market history, there have been far fewer successful market timers than successful stock selectors, and it is not clear whether even the few successes that can be attributed to market timing are more attributable to luck. This can be attributed to the fact that it is far more difficult to gain a differential advantage at market timing than it is at stock selection. For instance, it is unlikely that one can acquire an informational advantage over other investors at timing markets, but it is still possible, with sufficient research and private information, to get an informational advantage at picking stocks. Market timers contend that they can take existing information and use it more creatively or in better models to arrive at predictions for markets, but such approaches can be easily imitated, and imitation is the kiss of death for successful investment strategies.