Richard Pzena’s Q3 Letter
Using short-term volatility when making long-term investment decisions can lead to dubious results. A strong case for equities emerges when using consistent time horizons for the investor with long-dated liabilities.
Despite the fact that equity markets have continued their recovery from the depths of the Global Financial Crisis in early 2009, investors can’t stop looking nervously over their shoulders. While some measure of caution is always good, investors – both individuals and institutions – have become so loss-averse that they are increasingly putting themselves in the position of micro-managing volatility and making increasingly more short term tactical changes to their portfolios. The wide acceptance of short–term volatility as a measure of risk and the Sharpe ratio as a way of comparing risk and return has only served to increase the focus on the short-term. Witness the rise in the use of hedge funds or risk parity strategies within the portfolios of many investors. These strategies generally accept lower returns on the basis that the risk adjusted returns (i.e., Sharpe ratios) are actually higher. We have always maintained that short term volatility as a measure of risk in an asset class that is long-term by its very nature (i.e., equities) is inappropriate. In this quarter’s commentary we further examine the question of what makes for long-term investment success and consider the negative consequences of aiming for reduced short-term volatility. We conclude that investors with long-term investment horizons are best served by using investment data that matches the time horizon of the liabilities that the investment assets support.
Richard Pzena: Matching Time Horizons is Critical
In their most recent institutional investor study, Pyramis Global Advisors notes that “To keep pace with today’s market investors are shortening their time horizon and managing their portfolios more dynamically…..trying to move to a more real-time process.” They are doing so both in terms of their use of risk measurement metrics as well as with the greater use of short term investment management tools. The survey provides additional interesting insight on this point. In the response to the question “What changes have you made to your risk-management process in the last three years (post 2008)?”, nearly half of their respondents indicated they are looking at their risk measurements more frequently while over 40% said they had moved to being more tactical in their asset allocations.
The question is, will more frequent decision making and focus on short-term risk management really contribute to the process of meeting the long-dated liabilities? Currently the average duration of U.S. pension liabilities is seventeen years (Figure 1), making for a very long investment time horizon. Even the average retiree today has a life expectancy of eight years and therefore needs assets to be working with that time horizon in mind.
A recent academic study by Kinlaw, Kritzman and Turkinton published by MIT’s Sloan School of Business documents the erroneous conclusions that can be reached by using short-term volatility metrics in the Sharpe ratio calculation. In their paper, they compared the use of monthly volatility with the use of triennial (every three years) volatility and observed significant differences in the Sharpe ratio. They cited one example of a hedge fund that ranked in the top quartile on the basis of monthly volatility, but in the bottom quartile on the basis of triennial volatility. We would go further and argue that even triennial volatility isn’t a long enough measurement period. Since even retirees have an average duration of eight years for their liabilities, we should be interested in comparing the expected return for an investment strategy to the variability of that return for the same period. By using a shorter measurement period, investors may be accepting lower returns without truly mitigating their risk. The main point here is that short-term volatility is an unreliable indicator of longer term volatility, yet this is linkage is implicitly assumed when we use short-term volatility in the calculation of risk..
In 2011 we published a white paper on this topic. The conclusions still hold true: the volatility of returns diminishes as the holding period increases. Much of the daily or monthly volatility of a portfolio disappears as the vagaries of short term price movements offset each other in the long term and the frequency of loss decreases substantially.
Interestingly, when we compare the volatility of equities to the volatility of hedge funds, we reach some startling conclusions. The volatility of equities falls sharply as holding period increases. But for hedge funds, that is not the case, as shown below:
The history of value in periods following market corrections is one predominantly of outperformance, in many cases by a significant margin. During our study period, value outperformed in thirteen of fourteen periods post-correction (Figure 1), leading to value stocks adding almost 2.5% per annum of excess return over the S&P 500 for the entire fifty-four year period.
We speculate that hedge funds structure themselves to reduce short term volatility, because that is what is in demand. But that volatility can be reduced in traditional equity or fixed income portfolios simply with the passage of time. Most “low–vol” products come with higher fees and lower returns than simple equity products. Paying high fees or accepting lower returns appears dubious in spite of the attractive Sharpe ratios. The table below brings the longer term volatility and returns together.
Measuring from the peak of the market during the height of the internet bubble, the worst possible starting point for equity returns, hedge funds earned 4.7% per annum compared to global equities of 4.3%. But from the market’s trough, hedge funds have massively underperformed equities. If we speculate that the average experience of a long-term investor is going to be somewhere in between the peak and the trough, the only case for hedge funds is if the risk is actually lower. The historical experience, however, is that hedge fund returns were roughly half that of equities with similar long-term volatility. So the measurement period of that risk becomes critical. For the long term investor, the answer is easy – equities have a clear advantage. For a short term investor or a trader, there is clearly a tradeoff to be made.
Since 2008, investors have increasingly focused on short term metrics in making investment decisions, and their desire to reduce risk has driven them to implement various “low volatility” strategies and tactics. Unfortunately all of these actions revolve around the use of short-term volatility data. This seems to be fundamentally at odds with the basic underpinnings of what makes long term investing so successful. The better choice seems obvious: use long-term investment data that better fits with the time horizon of the liabilities that the investment assets support.
Past performance is no guarantee of future results. The historical returns of the specific portfolio securities mentioned in this commentary are not necessarily indicative of their future performance or the performance of any of our current or future investment strategies. The investment return and principal value of an investment will fluctuate over time.
The specific portfolio securities discussed in this commentary were selected for inclusion based on their ability to help you understand our investment process. They do not represent all of the securities purchased, sold or recommended for our client accounts during any particular period, and it should not be assumed that investments in such securities were, or will be, profitable.