One of my first articles for Seeking Alpha demonstrated a very simple momentum strategy that had created higher risk-adjusted returns than owning stocks outright over very long time intervals. The strategy examined the most widely followed equity benchmark, the S&P 500 (SPY, IVV), and the most oft used fixed income benchmark, the Barclays Capital U.S. Aggregate Index (AGG), and dictated that investors purchase the index that had performed the best in the trailing one month and hold that index forward for the next one month. Over thirty-five years, this monthly switching strategy produced a return stream that had roughly the same average return as passively owning the S&P 500, but with only 2/3 of the risk as seen through the lower standard deviation of returns.
Recently, I updated the data for the first six months of the year in the comments section of the original article. The momentum strategy continued to perform well in the first half of 2012, generating a return almost equal to owning stocks outright, but with less than half of the volatility and only one down month. Performance of the momentum strategy was driven by being long equities during the February and March rallies, but missing the May sell-off because bonds had outperformed in April.
(click to enlarge)
Seeking Alpha founder and CEO, David Jackson, recently commented on the original article and raised the question about how this strategy would have performed with an emerging market stock index included.
The most commonly used emerging market stock index is the MSCI Emerging Markets Index, which has available data back to the end of 1987. The two largest emerging market stock exchange traded funds: Vanguard MSCI Emerging Markets ETF (VWO) and iShares MSCI Emerging Markets Index (EEM) both utilize this reference index. Examining a three asset portfolio that purchases only the Barclays Agg, the S&P 500, or the MSCI Emerging Markets index based on which performed best in the trailing one month yields stunning results.
Below is performance data for the three indices and the momentum portfolio since the beginning of 1988 and a graph displaying the cumulative return profile of each. The cumulative return profile is the performance of $1 invested in each index and the momentum portfolio at the beginning of 1988. By the end of 2011, the momentum portfolio is 3.12x larger than the emerging market stock index, 3.99x larger than the U.S. stock portfolio, and 6.23x larger than the U.S. bond portfolio.
The momentum portfolio generated this level of outperformance over the twenty-three year sample period while exhibiting only 70% of the volatility of the emerging market index. The momentum portfolio produced substantially greater risk-adjusted returns than the S&P 500 index. To demonstrate this outperformance, imagine you could obtain zero-cost leverage on the S&P 500 index and held a position that was 114.8% invested in this index. This level of leverage would boost the riskiness of your portfolio to that of the momentum portfolio (17.17%, the standard deviation of the momentum portfolio divided by 14.95%, the standard deviation of the S&P 500 equals 1.148, the size of the leveraged portfolio), but you would only earn on average 14.8% (the leverage component) multiplied by 10.64% (average return on the S&P over that period) equals 1.58% more per year than the un-levered S&P 500 portfolio. With the momentum portfolio you would still earn an extra 5.22% per year over the equally risky levered S&P 500 portfolio, demonstrating the sizeable alpha available for similar risk through the momentum portfolio.
The beauty of this strategy is that it is so simple to construct. For investors purchasing stocks on a subscription plan as many do in their retirement vehicles, they will simply rotate their purchases towards the index which had outperformed in the last month. In this sample period, investors would have bought emerging market stocks in 133 months, the bond index in 87 months, and the S&P 500 portfolio in 67 months.
With the S&P 500 being the index utilized least often, I also examined a two-asset portfolio of just the emerging markets index and the domestic bond index. The resultant returns were actually slightly better than when the S&P 500 was including as a portfolio choice. The momentum portfolio invested in emerging market stocks in 152 months and the domestic bond portfolio in 135 months, producing an average annual return of 17.99% and an annualized standard deviation of 17.23%. Investors trying to time their allocation to emerging market equities would be well served to examine the returns of the recent past.
The chart below lists the performance by calendar year of the two-asset momentum portfolio. The year 1988 is excluded because the momentum portfolio was formed in February based on January results, so a full year of performance is unavailable. From 1989 through 2007, the momentum portfolio produced a negative performance only once, a feat even the lower risk bond index could not replicate. However, three of the last four years would have produced negative returns, including a markedly bad performance in 2010 when both indices posted positive returns.
(click to enlarge)
Despite this recent rough stretch for the two asset momentum portfolio, I believe that including emerging market investing in a momentum portfolio has value and can help guide asset allocation decisions. While backcasted results are no guarantee of future performance, momentum strategies have been shown to have value across asset classes and time periods. The relationship between emerging market stocks, domestic bonds, and domestic stocks will continue to evolve in ways that differ dramatically from the past twenty-three year sample period. Given the low interest rate environment, positive returns from monthly allocations to the bond component are unlikely to meaningfully increase absolute performance. This momentum strategy is meant to help investors miss periods that are very negative for their portfolios (either of these strategies would have generated positive returns from owning bonds in the fourth quarter of 2008 when domestic and emerging market stocks returned -22% and -26% respectively) while remaining invested in equities when returns are very strong.
Author's Note: For readers interested in examining the return data for the three indices and the momentum portfolio, the data is included below.
(click to enlarge)
Disclosure: I am long SPY.
Additional disclosure: I have frequently owned VWO and AGG, and will do so again. Currently, I am tactically underweight emerging market stocks and U.S. investment grade bonds in favor of U.S. stocks, but will again add to the former asset class in the near-term.