Michael Kitces   |   December 19, 2011

Readers who visit the Pinnacle Advisory Group website (www.pinnacleadvisory.com) will note that a paper by Solow, Kitces, and Locatelli, entitled “Improving Risk Adjusted Returns Using Market-Valuation-Based Tactical Asset Allocation Strategies,” was published in the December issue of the Journal of Financial Planning (JFP) – the most distinguished journal in the profession. The paper itself is a technical study that may be difficult to read, so this summary is intended for those who might not care to wade through the academic and statistical details. The basic conclusion of the paper was this: By reducing exposure to stocks when their prices are excessive relative to the profits they produce and increasing exposure to stocks when prices are low relative to their earnings, it is possible to systematically improve long-term returns.

In order to understand the impact of changing exposure to stocks based on their valuation, we first looked at the historical returns associated with stocks depending on whether they were extremely overvalued, extremely undervalued, or somewhere in between. We measured the relative over- or under-valuation of stocks by looking at a popular measure called the P/E (Price/Earnings) ratio, which measures the value of stocks compared to the earnings they produce. In order to smooth out the data and focus on long-term returns, we used a particular version of the P/E ratio that compares the price of the markets to the average level of earnings over the past 5 years. If this ratio was in the top 10% of its normal range, we assumed markets were overvalued; if the ratio was in the historical bottom 10%, we assumed markets were undervalued; the rest of the time, we simply assumed stocks were “close enough” to average value to simply buy and hold them.

Our results in this portion of the study showed that, on average, looking at market valuation really does matter. Although there were exceptions in some years, the average return of stocks in the years that followed high valuation (i.e., “overvalued”) environments were significantly lower than the overall long-term average return of stocks. Similarly, when stocks were undervalued, the following years on average produced superior returns.

Based on this framework – that stocks predictably produce below-average returns in overvalued environments and above-average returns when they are undervalued – we set forth a series of rules for tactically changing the asset allocation of a simple portfolio of stocks and bonds. For instance, instead of always holding a portfolio with a 50/50 mix of stocks and bonds, we assumed that the investor would make a tactical shift to increase stocks to 60% in undervalued environments, or reduce them to 40% when they are overvalued; in other words, the portfolios could range from 60/40 to 50/50 to 40/60. (We also analyzed more dramatic shifts.) We then evaluated how a portfolio making such tactical shifts would have performed during any particular 30-year period over the past 85 years, given the actual historical returns and fluctuations of the markets.

The results revealed that the tactical portfolios achieved higher long-term returns than the buy and hold portfolio, consistent with our hypothesis. Initially, the results suggested that these results were only achieved due to an increase in the riskiness of the portfolio (as measured by its volatility). However, further analysis revealed that the increased volatility was only upside volatility, while volatility to the downside was reduced. In other words, the portfolio was simultaneously achieving better returns in favorable stock environments by overweighting, and not losing as much money in more dangerous overvalued stock environments thanks to underweighting. In fact, the results showed that the strategy was so effective, adjusting equities up or down by 30% (e.g., from 20/80 to 80/20) was even more beneficial than just adjusting from 60/40 to 40/60, and it was slightly more effective to go all the way from 0/100 to 100/0!

Our analysis also considered the impact of taxes, based on the current tax laws, since additional tactical trades will cause gains to be recognized slightly more rapidly than a basic buy and hold approach. Nonetheless, the results showed that the tactical portfolios remained superior to the buy and hold portfolios, even in light of ongoing taxation, due in part to the fact that the tactical approach is still only designed to generate an average of two trades (over- or under-weighting equities) per decade.

Our study concludes that even a simple rules-based method for tactically changing the asset allocation of a portfolio of stocks and bonds based on valuation is capable of beating a buy and hold portfolio strategy in terms of absolute and risk-adjusted returns. Notably, the tactical approach contained in this study is somewhat simplified; in reality, Pinnacle Advisory Group actually uses complex quantitative (rules-based) and qualitative (judgment and experience-based) methods for tactically managing client accounts.  However, for those who claim that tactical asset allocation is no more than a “guessing game” utilizing market timing strategies that cannot succeed in efficient markets, we hope this study advances the notion that active portfolio management can in fact lead to superior risk-adjusted returns, and even an outright increase in long-term return.