A related paper has been added to:

#118 – Time Series Momentum Effect

Authors: Till

Title: What are the Sources of Return for CTAs and Commodity Indices? A Brief Survey of Relevant Research

Link: http://www.oxfordstrat.com/coasdfASD32/uploads/2016/03/Sources-of-Return-for-CTAs.pdf

Abstract:

This  survey  paper  will  discuss  the  (potential)  structural  sources  of  return  for  both  CTAs  and  commodity  indices  based  on  a  review  of  empirical  research  articles  from  both  academics  and  practitioners.  The  paper  specifically  covers  (a)  the  long-term  return  sources  for  both  managed  futures  programs  and  for  commodity  indices;  (b)  the  investor  expectations  and  the  portfolio  context for futures strategies; and (c) how to benchmark these strategies.

Notable quotations from the academic research paper:

"In the academic literature, one can find strong evidence – historically at least – for there being persistent returns in futures programs due to momentum, roll yield, and also due to rebalancing. This is actually the case across asset classes, and not just for commodity futures contracts.

The AQR authors theorised that “price trends exist in part due to long-standing behavioural biases exhibited by investors, such as anchoring and herding, as well as the trading activity of  non-profit seeking participants, such as  central banks and corporate hedging programs.” Assuming these factors continue, the long-term profitability from momentum strategies might also continue, and not just be a matter of history.

“However, the … strategy also exposed investors to large losses … during both [historical] periods,” noted the Federal Reserve Bank of Chicago paper (Chabot et al. (2014)).  Interestingly,  “[m]omentum  …  [losses]  were [apparently]  predictable”.  In  both  historical  periods,  losses  were  “more  likely  when  momentum recently performed well.” For the 1867 to 1907 period, losses were more likely when “interest rates were relatively low.” And for the 1927 to 2012 period, losses were more likely when “momentum had recently outperformed the stock market”. Each of these periods were “times when borrowing or attracting return chasing  ‘blind  capital’  would  have been  easier.”  The authors argue that the periodic large losses, associated with the strategy plausibly becoming too popular, “play an important role in sustaining” the momentum strategy’s historical returns.

In  addition to momentum,  the empirical literature also documents that “roll yield” can be considered a structural source of return, at least over long periods of time. For example, Campbell & Company  (2013) described a proprietary trend-following benchmark, in which they calculated returns from 1972 through November 2012, and which included a selection of equity, fixed income, foreign exchange, and commodity markets. Over this 40-year period, approximately half of the benchmark’s cumulative performance was due to spot return, and the other half was due to roll yield. Over long horizons, the roll yield is important mainly for commodity futures contracts. This is because of another structural feature of commodity markets: mean reversion. If a commodity has a tendency over long enough timeframes to mean-revert, then by construction, returns cannot be due to a long-term appreciation (or depreciation) in spot prices. In that case, over  a  sufficient time frame, the futures-only return for a futures contract would have to basically collapse to its roll yield. Can we observe this historically in commodity futures markets? The  answer is essentially yes.

The mean-reversion of commodity prices can also have meaningful consequences for returns at the portfolio- or index-level. Specifically, this feature is at the root of an additional source of return, quite separate from trends in spot prices or the potential persistence of curve-structure effects. That potential additional source of return is the return from rebalancing. Erb and Harvey (2006) discussed how there can be meaningful returns from rebalancing a  portfolio of lowly-correlated, high-variance instruments. The rebalancing effect was explained Greer et al. (2014), as follows: “[A] ‘rebalancing return’ … can naturally accrue from periodically resetting a portfolio of assets back to its strategic weights, causing the investor to sell assets that have gone up in value and buy assets that have declined.”

One caveat is that one’s holding period may have to be quite long term in order for these return effects to be apparent. However, even structurally positive returns may be insufficient to motivate investors to consider futures products. A CTA (or global macro) investor may require that the program’s return profile is also long-options-like; and an institutional investor will expect that a commodity index will provide diversification for a stock-and-bond portfolio. The paper also noted that how these programs are benchmarked will depend on whether a futures program is considered a beta, an alternative beta, pure alpha, or well-timed beta. This paper correspondingly provided recommendations for benchmarks for each of these types of investment exposures."

Are you looking for more strategies to read about? Check http://quantpedia.com/Screener

Do you want to see performance of trading systems we described? Check http://quantpedia.com/Chart/Performance

Do you want to know more about us? Check http://quantpedia.com/Home/About