Did you know that modern portfolio theory has been incorrectly applied by the industry for the past 60 years?
The foundation of investment education for CFP certificants is modern portfolio theory, which gives us tools to craft portfolios that effectively balance risk and return and reach the efficient frontier. Yet in his original paper, Markowitz himself acknowledged that the modern portfolio theory tool was simply designed to determine how to allocate a portfolio, given the expected returns, volatilities, and correlations of the available investments. Determining what those inputs should be, however, was left up to the person using the model. As a result, the risk of using modern portfolio theory – like any model – is that if poor inputs go into the model, poor results come out. Yet what happens when the inputs to modern portfolio theory are determined more proactively in response to an ever-changing investment environment? The asset allocation of the portfolio tactically shifts in response to varying inputs!
The evolution of the industry for much of the past 60 years since Markowitz’ seminal paper has been to assume that markets are at least “relatively” efficient and will follow their long-term trends, and as a result have used historical averages of return (mean), volatility (standard deviation), and correlation as inputs to determination an appropriate asset allocation. Yet the striking reality is that this methodology was never intended by the designer of the system itself; indeed, even in his original paper, Markowitz provided his own suggestions about how to apply his model, as follows:
To use [modern portfolio theory] in the selection of securities we must have procedures for finding reasonable [estimates of expected return and volatility]. These procedures, I believe, should combined statistical techniques and the judgment of practical men. My feeling is that the statistical computations should be used to arrive at a tentative set of [mean and volatility]. Judgment should then be used in increasing or decreasing some of these [mean and volatility inputs] on the basis of factors or nuances not taken into account by the formal computations…
…One suggestion as to tentative [mean and volatility] is to use the observed [mean and volatility] for some period of the past. I believe that better methods, which take into account more information, can be found.”
- Harry Markowitz, “Portfolio Selection”, The Journal of Finance, March 1952.
The whole article, The Rise of Tactical Asset Allocation, by Michael Kitces is a great summary of the problems with the way that modern portfolio has been applied. Of course, even if the industry had taken Markowitz’ advice and tried to forecast the inputs of standard deviation, covariance, and expected return they would have run into an entirely different problem—without a crystal ball, trying to forecast those inputs is no better than simply taking the historical means!
As Kitces correctly points out, there continues to be rising demand for an alternative approach to asset allocation. At Dorsey Wright, we espouse a trend following approach to asset allocation. Specifically, we allow relative strength to determine how a multi-asset class portfolio will be allocated. It is flexible, pragmatic, and it works. Try talking to your clients about it and don’t be surprised if they agree that it makes a lot of sense.
HT: Abnormal Returns