Tuesday, June 9, 2020

Turning points kill trend-following performance


Markets trend higher and trend-followers should make money. Markets trend lower and trend-followers should make money. The transitions or turning points are the problem. There will be periods of giveback and execution delay while the new trend signals are found. If there are more turning points or reversals, returns will suffer. This enemy of trend-followers is all well-know, but a new paper documents the cost of turning points. See "Breaking Bad Trends" by the researchers at Research Affiliates and Campbell Harvey of Duke University. The elegance of this paper is with its simplicity. 

The return impact of turning points is devastating for exploiting trends. Using a simple approach of comparing 12-month to 1-month return direction as an indicator of a turning points, the researchers find that as the number of turning points per asset increase the performance from a trend strategy declines. When the number of turning points for an asset reaches six within a year, the median Sharpe ratio falls below zero. This is intuitive but can be used as a good starting point to explain why trend-followers may lose money.  


The second graph shows the portfolio performance versus a weighted average number of asset turning points per year. There is a negative linear relationship between portfolio returns and turning points by year. The more recent periods have shown more turning points, on average, for markets. This performance pattern is similar to what has been found with trend-following managers. Asset turning points have become more frequent in spite of the fact that many markets have become less volatile in the post-GFC period. 


Their final graph is surprising. The static trend strategy seems to be independent of volatility. If you believe this graph, the work was done carefully and well-documented, the idea that trend following is a long volatility strategy is called into question. It does not matter what is the volatility environment. The turning points matter more.

The switching between long and short momentum trends has increased in recent times and has an appreciable impact on static trend-following. The paper offers further analysis on how to adapt to turning points through a dynamic approach to adjusting to changing trends. This is a fruitful area for further research. The challenge for trend followers is to find ways to adapt to these turning points. In fact, investors should spend less time on how trends or found but how managers adapt to the turning points.

No comments: