ETF Complete Strategy Insights: Momentum in the ETF Models (Part 2)

James Kimball | August 4, 2019

The ETF Complete model closed the week down -2.8% compared to the SPY which closed down -3.1%.

The SPY started the week near its all-time high as earnings came in better than some had feared. However, the FED announced a quarter-point rate-cut on Wednesday followed by an indication from the President that the trade war truce was off again, sending markets lower, closing back near its 50-DMA on Friday.

Stay tuned for the daily updates and log into the website to see holdings and additional performance data.

image

This Week's Strategy Lesson: Momentum in the ETF Models (Part 2)

image

Last week we started this series on momentum looking at some of the characteristics of typical trades in the ETF models. Many of the best trades were at all-time or multi-year highs. But even the trades that weren’t on new highs had one thing in common, we tend to buy them on strength.

There are a lot of ideas about what works in trading and markets. Many will swear by “value” or “growth” investing or fundamental analysis or some key metric like the price-to-earnings ratio, book value, or year-over-year earnings or dividend growth. Still others might ignore these styles and metrics and focus exclusively on technical analysis and market timing. Each of these have their place in the trader’s toolbox and have proven useful with different applications or in different market conditions.

Is there one technique or approach that has an edge? James O’Shaughnessy’s best-selling book, “What works on Wall Street” is one of the best places to start researching this topic.

In his book, he takes a wide-view approach, looking at large amount of data from thousands of stocks between 1927 and 2009 and surveying multiple perspectives including fundamental and value investing to pure momentum.

As would be expected, each method has it merits and can work better or worse in different market conditions, however, buying stocks that have had the best price appreciation over certain periods proved to be the top performing and most persistent strategy in his large dataset.

For instance, O’Shaughnessy looked at ranking stocks with a simple six-month return and found that buying the top decile (top 10% of rankings) of stocks based on six-month return, over the entire sample period of 82 years resulted in an average annual return of 17.6% versus 13.1% for all stocks—a nearly 35% improvement in annualized return.

Furthermore, the top decile beat the all stocks group in 87% of all five-year rolling periods and 98% of all ten-year rolling periods (though it was not without some serious drawdowns and had several multi-year periods where it underperformed).

image

The chart above shows the excess annualized returns of a portfolio of the top decile stocks rated by six-month momentum versus the all stocks group from January 1927 to December 2009, taken from “What Works on Wall Street,” 4th edition, 2012 Page 411.

This is a strong result and the “edge” from this strategy is persistent and survives decade after decade of stock data. However, in this form, we still do not have a completely viable or trade-able strategy. O’Shaughnessy is looking at big data. To get precisely these results would require potentially hundreds or thousands of trades to put you in and keep you in (through rotation) the top-rated stocks.

You also wouldn’t be immune from some serious drawdowns, as this strategy at one point over its 82-year history experienced a 78% drawdown—although that was less than the 89% drawdown in the Dow Jones Industrial Average from September of 1929 to July of 1932.

Many of the elements of our ETF strategies are designed to address these issues. We trade representative ETFs that can give us exposure to the top performing sectors or assets classes without requiring hundreds of trades or complicated rebalancing. Our Trend Strength Indicator (TSI) calculation works better than simple six-month or twelve-month price appreciation calculations. And we have short exposure and rules related to negative TSI values that are designed to reduce drawdowns.

We have now seen both some of our own ETF trades that demonstrate the power of buying strength as well as some broad statistical data to back this up. But what fundamental or real-world phenomenon might be giving momentum this structural advantage? Next week we will wrap up this series trying to answer this question.