Let me start by saying that I believe a low-cost, passive investment strategy based on the tenet of diversification is a prudent starting point for building durable and efficient portfolios. The elephant in the room though with passive investment strategies is valuation. Ignoring near-term market conditions, including valuation, is neither reasonable nor prudent in my opinion. Taking a line from Jeremy Grantham's latest quarterly note:
Stocks at fair value are less risky than stocks trading 30% above fair value because the expensive stocks give you the risk of loss associated with falling back to fair value … When an expensive asset falls back to fair value, subsequent returns should be assumed to be normal, which means that the loss of wealth versus expectations is permanent.
Consider current valuations in U.S. Treasuries - how does a passive asset allocation strategy contend with the upcoming regime change in interest rates and reconcile rebalancing a portfolio into a fixed income bubble? Investment allocation decisions should not be made in a vacuum and once information is assimilated to make informed decisions - then acting, in what I would consider a prudent manner, can no longer be a passive exercise.
If one argues that valuations are accounted for in a strategic asset allocation by using some form of (mean-variance) optimization, especially one based on future return expectations, correlations, and variances -- then passive shifts to active and strategic blurs with tactical.
The problem is that extending beyond some naive or static allocation process brings us a grab-bag of active or tactical strategies, and unfortunately many of these strategies fail to beat the comparable passive indexes. Well known passive investment strategists highlight the performance of a low cost balanced index fund such as Vanguard's Balanced Index Fund (VBINX) as proof that you shouldn't waste your time with active strategies.
Morningstar recently updated its rather unflattering, but not unexpected, report that showed when it reviewed the performance of 163 tactical funds relative to Vanguard's Balanced Index Fund, a balanced fund was hard to beat on several risk and return measures. While the time period is somewhat dubious, the overriding point is not that the grab-bag of tactical funds Morningstar elected to review underperformed passive indexes but that a majority of active management strategies underperform passive indexes.
Still - how to align a low-cost, efficient investment strategy while acknowledging that the elephant in the room has a good chance of squashing returns if you don't pay attention to current valuations? I argue that passive asset allocation strategies can coexist without explicitly measuring valuation by focusing on lowering volatility. Two ways that you can improve the risk-adjusted returns of a core strategic asset allocation portfolio include:
- Target low-volatility tilts for each asset class
- Use a quantitative model designed to avoid market volatility
As for low-volatility tilted asset classes - historically this have been done through dividend-oriented or value-based tilts. However, new investment options are appearing that allow investors to target groups of stocks that show lower volatility relative to the market. Earlier this week, I had the opportunity to attend a presentation by Brendan Bradley from Acadian Asset Management on his research on the low-volatility anomaly and how a managed volatility portfolio can achieve similar returns to the broad market while lowering its risk profile.
Low-volatility concepts are now available to the larger investment community given that index providers such as MSCI and Standard & Poor's are creating low volatility indexes. PowerShares has launched several low volatility ETFs based on these indexes, including the S&P 500 Low Volatility Portfolio (SPLV), S&P International Developed Low Volatility Portfolio (IDLV), and the S&P Emerging Markets Low Volatility Portfolio (EELV).
While lowering volatility seems reasonable, it is the utilization of a quantitative or mechanical timing model that brings with it the most negative connotations. Let's be clear, this is working with time-series momentum and not relative momentum. I have benefited from the work of Mebane Faber (his website) and I have utilized his research to build my own processes. The concept is clear - deploy a quantitative strategy as an active risk management strategy to lower the volatility of an already diversified portfolio.
While the following figure is not the easiest to read, it shows the quantitative methodology of a simple trend system that can help avoid volatility in the market. Using the history of the S&P 500 SPDR ETF (SPY), Figure 1 shows that when the volatility of the S&P 500 ETF is above its average, the ETF tends to be priced below its 10-month moving average.
Figure 1: Intermediate Trend Relationship with Volatility
Click to enlarge.
Mr. Faber has presented several data sets that show risk and return metrics when various asset classes are above or below their intermediate trend line. In general, risk and return metrics improve materially when an asset class is trading above its intermediate trend, whether that is the 200 Day SMA or the 10 Month Moving average.
To conclude with a final Grantham quote from his quarterly letter:
"The cardinal rule is to not underperform in bear markets.
The combination of an efficient portfolio along with an active risk management strategy should go a long way in sticking to that cardinal rule. None of us know where the market will be tomorrow and to earn a return we have to invest in risk assets. While passive asset allocation strategies are hard to improve on -given market inefficiencies including periods of gross-asset class mispricing, I believe there are still intelligent ways to improve investment performance. Lowering volatility improves returns by reducing volatility drag both from a cross-sectional perspective and from a market timing perspective.