Trend-following - A solution but not a panecea

 

"CTA claims of being superior to other forms of risk mitigation are exaggerated because (1) the past performance has been dominated by persistent long positions in bonds over a decades-long period in which interest rates declined to effectively zero, and (2) the convexity with respect to equity moves is less effective than direct tail hedging." See Can an Allocation to CTAs Move the Needle?

I disagree.

If the trend in bonds is for lower yields trend-followers will be long bonds. If the trend for bonds is for higher yields, trend-followers will be short bonds. There is no angst. There is no doubt. A trend is a base case, and it is indifferent to the rate regime. Less money was made in shorting bonds over the last 40 years because there were fewer short bond opportunities.

Yes, the convexity with respect to equity moves is less effective than a direct tail hedge, but the cost of being wrong is less. If you know that equities are going lower, you don't need a hedge; you need a sell order. A tail hedge needs to get the direction, the magnitude of the move, and the timing right. An equity index trend moves only needs to get the direction correct.

The more important issue with this paper is that small allocations to CTAs will not have an appreciable impact on return or volatility reduction and holding a broad portfolio of CTA would have had strong periods of underperformance during the 2010 decade and the 1990's. The numbers do not lie. Trend-following will work but the impact is a function of the size of exposure. Trend-following like all strategies will fall in and out of favor. The periods of underperformance can be long, but that is part of the cost of buying convex strategies. There must be large moves to make these strategies attractive.