The more time I spend in the real world of finance, the more I question the rationale of textbook theories.
Recently, I have found myself realizing the inherent conflicts of generating positive alpha employing money management practices consistent with modern portfolio theory.
Many of us, professional financiers or not, are likely familiar with the relationship between risk and reward when it comes to investing. More risk should be accompanied with a larger potential reward. Modern portfolio theory attempts to maximize the level of return in a portfolio of stocks for a given level of risk. In addition to assuming markets are efficient–part of the theory I won’t dig into in this post–the measure of risk and the construct of a portfolio following these principles is, in my opinion, alpha prohibitive.
Risk, as defined by modern portfolio theory, is the standard deviation of portfolio returns. This tells an investor how far returns tend to move around its average, either above or below. Diversification is the tool used within this framework to minimize the level by which returns deviate. It states that proper diversification can eliminate risks specific to individual companies by pairing them with assets that are less, or even negatively, correlated.
This means if stock ABC is up 10% then stock XZY, which is less exposed to the same economic forces of stock ABC, may only be up 2%. If it is perfectly negatively correlated, the preferred but rare relationship, than stock XYZ would be down 10% exactly offsetting the movement, or deviation, of ABC.
If portfolios are constructed in such a way to remove firm-specific risks by offsetting them with less correlated assets it becomes increasingly difficult to generate alpha. This issue is magnified when measuring risk-adjusted returns relative to a benchmark with other metrics such as the information ratio. This is maximized when a portfolio outperforms its benchmark, or generates alpha, with a lower standard deviation of returns than the benchmark.
If we strip out the complicated formulas used to measure the maximization of return per level of risk and focus on how minimizing risk relative to a benchmark impacts the change in portfolio value, we see that it would result in benchmark-like performance. The best way to minimize standard deviation relative to the benchmark is to hold a portfolio of stocks that behave like this yardstick, which also means we are left with benchmark like returns. This would produce little to no alpha.
While the alternate of a more concentrated portfolio may or may not be feasible for all investors depending on their individual situations, I prefer it to a well-diversified strategy as defined by textbook theories. Risk, which can arguably be best defined as the chance of permanent loss of capital, can be managed within this approach by selecting high-quality companies. This includes those with exceptionally strong balance sheets and earnings with a large cash component, among other factors, that are trading large discounts to its true value based on reasonable profit growth assumptions.
With this method we can be left with a portfolio full of companies we feel have the best upside potential regardless of the correlation of their returns. While it may lead to larger swings in portfolio value due to its concentrated nature, it has a much better chance at delivering alpha.
*Alpha – A measure of performance on a risk-adjusted basis. Alpha takes the volatility (price risk) and compares its risk-adjusted performance to a benchmark index. The excess return relative to the return of the benchmark index is a fund’s alpha.
No strategy assures success or protects against loss. The opinions voiced in this material are for general information only and are not intended to provide specific advice or recommendations for any individual. To determine what is appropriate for you, consult a qualified professional.
