Bonds are held in the portfolio as a diversifier, risk inhibiter, and income generator. However, with low interest rates, it is only a matter of time until we see the "bond bubble" burst as interest rates will rise and bond values will fall. Assuming this projection has a high probability of occurring over the next few years, how can investors reduce portfolio risk without holding a high percentage of bonds in the portfolio?
To explain these models, we borrow and modified ideas articulated in detail in Mebane T. Faber and Eric W. Richardson's book, The Ivy Portfolio. Using their simple 10 asset class portfolio, the bond (BND) ETF is removed and replaced with a mid-cap blend ETF, VO. TIPS (TIP) is retained in the portfolio as a holding receptacle when an equity ETF is sold even though it is a type of bond ETF.
In the following screen shot the 10 ETF portfolio is analyzed using Geoff Considine's Quantext Portfolio Planner (QPP) software. The projected return over the next six to twelve months is 8.9% annualized when the S&P 500 is projected to be an annualized 7.0%. Portfolio uncertainty is high for this portfolio as the projected standard deviation is 18.4%. The Diversification Metric (DM) is a meager 16% when our goal is to have DM reach 40% or higher. The low DM value is an indicator this portfolio is not all that well diversified.
With such a high projected standard deviation, a negative two sigma event will strip away over 35% of the value of this portfolio. This magnitude of losses is to be avoided. But how, without adding more bonds?
The crux of the problem is explained in the correlation matrix shown in the following screen shot. Most of the ETFs are highly correlated. Only TIP brings significant diversity to the portfolio. The two commodity ETFs, DBC and GSG, help to some extent, but not to the degree needed.
Faber and Richardson provide a quantitative system that will help ward off the potential two sigma losses. The system is as follows:
Buy Rule: Buy when monthly price > 10-month Simple Moving Average.
Sell Rule: Sell and move to cash when monthly price < 10-month Simple Moving Average.
I modified these rules by substituting a 195-Day Exponential Moving Average for the 10-month SMA. A detailed explanation is available at this site.
The portfolio is reviewed each month for a number of reasons. 1) We want to avoid the wash rule. 2) Whipsaws are reduced. 3) When using commission free ETF, short-term trading fees are avoided.
In addition to using the ETF price position and its 195-Day EMA, we add in another metric not included in the Faber-Richardson book. The added information is the Point and Figure (PNF) graph for each ETF.
As I was writing this article, the price of VB was very close to crossing from above to below its 195-Day EMA. If that were to occur on the day the portfolio was under review, a sell would be triggered for VB. The following screen shot is a current PnF graph of VB. Note the X's in the right-hand column, indicating recent price movement is trending upward. This graph acts as an emergency brake to selling the VB ETF too quickly. There will be times when the PnF graph confirms the Faber-Richardson sell trigger. When used in combination, fewer trading errors are likely to occur. Again, this is discussed in more detail at ITA Wealth Management.
(click to enlarge)
Several portfolios were selected as candidates for testing the risk reduction model as they were trailing either or both Vanguard's Total Stock Market Index (VTSMX) and a customized benchmark, the ITA Index.
Using the combination of moving averages and Point and Figure graphs provide investors with straight forward methods to curb large potential losses to a portfolio without exposure to a high percentage in bonds.