Improving Your Return Profile With A Simple Momentum Strategy

|
 |  Includes: AGG, SPY
by: Ploutos

Tired of the volatility of equities? Nervous about owning bonds against the backdrop of historically low interest rates? We all know that an allocation to bonds will reduce the variance of the returns of your portfolio over time, but what is the correct allocation? Why should the sizing of this allocation be based on some heuristic regarding age or years to retirement? Fortunately, exciting work in finance regarding momentum strategies may provide a solution.

Below I present a simple two asset portfolio iShares Barclays Aggregate Bond Fund (NYSEARCA:AGG) & SPDR S&P 500 ETF (NYSEARCA:SPY). These ETFs replicate the performance of the Barclays Capital U.S. Aggregate Index, the most commonly referenced fixed income benchmark, and the S&P 500 Index respectively. Beginning with the advent of this fixed income index in 1976, the data detailed here is a snapshot the historic monthly returns of the two underlying indices. (Full population of the data is available from the author upon request.) The "Momentum" column is a portfolio that chooses between bonds and equities based on which asset class performed better in the previous one month, and owns that fund for the next one month.

Month AGG SPY Momentum
January-76 1.95% 12.17%
February-76 0.75% -0.84% -0.84%
March-76 1.20% 3.37% 1.20%
April-76 0.56% -0.78% -0.78%
May-76 -0.91% -1.11% -0.91%
June-76 1.41% 4.43% 1.41%
July-76 1.17% -0.48% -0.48%
August-76 1.79% -0.18% 1.79%
September-76 1.38% 2.58% 1.38%
October-76 1.08% -1.86% -1.86%
November-76 2.42% -0.41% 2.42%
December-76 1.82% 5.61% 1.82%
Sample Data Time Series
January-11 0.12% 2.37% 2.37%
February-11 0.25% 3.43% 3.43%
March-11 0.06% 0.04% 0.04%
April-11 1.27% 2.96% 1.27%
May-11 1.31% -1.13% -1.13%
June-11 -0.29% -1.67% -0.29%
July-11 1.59% -2.03% 1.59%
August-11 1.46% -5.70% 1.46%
September-11 0.73% -7.20% 0.73%
October-11 0.11% 10.79% 0.11%
November-11 -0.09% -0.32% -0.32%
December-11 1.10% 0.86% 1.10%
Click to enlarge

When we annualize these monthly returns and the standard deviation of these returns, we get some extraordinary results. A portfolio that rotates between equities and bonds as proxied by the S&P 500 and Barclays Agg based on their trailing performance achieved nearly the same annualized return as a portfolio invested wholly in equities with only about 2/3 of the risk as defined by the annualized standard deviation of returns. Befitting any investor's goal of maximizing risk-adjusted returns, this is a powerful outcome indeed for such a simple concept and easily executable strategy.

Statistic AGG SPY Momentum
Average Return (Since 2/76) 8.27% 10.56% 10.53%
Standard Deviation 5.62% 15.30% 10.32%
Trailing 10-Yr Avg. Return 5.78% 2.83% 7.51%
Standard Deviation 3.70% 15.93% 9.94%
Trailing 5-Yr Avg. Return 6.51% -0.42% 8.05%
Standard Deviation 3.60% 18.90% 11.26%
Trailing 3-Yr Avg. Return 6.78% 13.78% 14.02%
Standard Deviation 2.82% 19.01% 10.84%
2011 Return 7.86% 1.24% 10.77%
Standard Deviation 2.35% 16.03% 4.42%
Click to enlarge

Plotting this momentum portfolio against varying weights of the combination of AGG and SPY demonstrates the efficiency of this strategy. Akin to Markowitz's efficient frontier, risk is defined by variance of returns and is plotted on the x-axis with total return on the y-axis. Visually, one can see that the momentum strategy over this long time horizon has produced a portfolio that generates excess returns relative to the expected level of risk.

Click to enlarge

Below are the annual returns of the three strategies. Equities are the most likely to outperform on a standalone basis with momentum and bonds equally likely to have the top performance. The ability of the momentum strategy to miss downturns (investing in bonds in 4Q08 in the wake of the Lehman bankruptcy) or ride the equity market when it is outperforming bonds (2003) is the key to its generation of risk-adjusted returns.

Year AGG SPY Momentum
1977 3.03% -7.16% -3.60%
1978 1.40% 6.57% 8.58%
1979 1.92% 18.61% 8.78%
1980 2.71% 32.50% 40.43%
1981 6.26% -4.92% 2.25%
1982 32.64% 21.55% 16.64%
1983 8.37% 22.56% 14.19%
1984 15.15% 6.27% 4.87%
1985 22.13% 31.73% 35.69%
1986 15.25% 18.66% 4.52%
1987 2.76% 5.25% 28.47%
1988 7.88% 16.61% 10.71%
1989 14.53% 31.69% 25.63%
1990 8.95% -3.10% 2.72%
1991 16.00% 30.47% 12.34%
1992 7.40% 7.62% 3.81%
1993 9.75% 10.08% 3.58%
1994 -2.92% 1.32% -3.12%
1995 18.48% 37.58% 29.60%
1996 3.61% 22.96% 26.11%
1997 9.68% 33.36% 11.41%
1998 8.67% 28.58% 32.70%
1999 -0.83% 21.04% 8.99%
2000 11.63% -9.10% -4.99%
2001 8.42% -11.89% -3.54%
2002 10.27% -22.10% -2.99%
2003 4.11% 28.68% 26.32%
2004 4.34% 10.88% 10.47%
2005 2.43% 4.91% -6.16%
2006 4.33% 15.79% 10.30%
2007 6.96% 5.49% 9.33%
2008 5.24% -37.00% -9.14%
2009 5.93% 26.46% 28.43%
2010 6.56% 15.05% 4.21%
2011 7.86% 1.24% 10.77%
Click to enlarge

Too often investors are too consumed with trying to find the appropriate mix of bonds and stocks in their portfolio when it is far easier to take our cues from the market's recent past. Given the increased liquidity afforded individual investors through exchange traded funds, building broad replicating portfolios of the most oft referenced stock and bond indices is far easier, making this strategy accessible for any investor.

Disclosure: I am long SPY.