Commodity roll yields are usually discussed in the context of optimising the risk and return of long only commodity indices, or explaining why commodity indices have lagged spot performance but the “commodity carry trade” can be a huge engine of alternative beta in its own right, and one that eclipses both long only commodity returns and other risk premiums. The strategy could hardly be simpler, which is why it should be dubbed alternative or exotic beta that is passively replicable as opposed to alpha that’s not – and the ease of execution leaves plenty of time for 3 hour long French lunches.
In a paper released last year, Devraj Basu and Joëlle Miffre at the Edhec Risk Institute, located opposite the airport in sunny Nice en Provence in the South of France, say that simply shorting contango and buying backwardation (each inferred from weekly CFTC Commitment of Traders Reports) for 27 commodities, with regular active rebalancing, would have earned you up to 11.8% per year, between 1992 and 2008 , compounding up to a 600% return over 16 years, when an equally weighted long only commodity portfolio averaged only 1.8% a year. Of course choosing 2008 – the nadir for commodities – as an end date flatters the comparison with long only, but there is no denying this is one monster of a risk premium, at least 8% ahead of risk free rates averaging around 3% over the period.
It’s extremely easy to leverage the currency carry trade, but without leverage it’s generally estimated at 4-5% a year (judging by this Thomson Reuters press release of February 8th 2010, which put 20 year venture capital returns at 17.4% well ahead of buyouts at 9.1%). Corporate or emerging market debt rarely comes in much higher over a full cycle. We all know the equity risk premium has disappeared over the past “lost decade.” The Dow Jones Credit Suisse hedge fund index (from 1994) has averaged near 8.5%.
Moving over to illiquids, you would need to have owned venture capital (or had the foresight to pick top decile managers for buyouts), to have matched the commodity carry trade return from private equity, naturally with far less liquidity; relatively boring buyouts only averaged 9.1% over 20 years, in absolute terms, trumping equities but still a smaller premium over risk free rates than equities. (Source: various)
Loves Risk Aversion
Whereas most carry trades have some correlation with risk appetite, the commodity carry trade has often performed best in periods of risk aversion – it had a great 2008. As such it has been a good diversifier for long biased commodity exposure or indeed risk assets in general. Does anyone have any idea why this is the “odd one out” of the carry trade family?
Attributing Returns and Correlation
Most of the return (78%) came from buying backwardation with 22% from shorting contango. However, from a diversification perspective either of the two strategies alone was 0.6 to 0.7 correlated to long only commodities. It is only the combination of the two that reduces the correlation coefficient to 0.06 (as you can see from the chart below created using data from Table 2 of the Basra and Miffre paper).
But is this investable ?
You might be able to synthetically isolate part or all of this risk premium by shorting a commodity index that rolls sequentially, and owning one that optimises its roll (but watch out for different baskets!) However it would seem less cumbersome to execute the strategy on a standalone basis – there may be indexes or systematic products dedicated purely and solely to capturing the commodity carry trade risk premium ? Trading commodity futures on margin could allow for substantial leverage to be applied to the strategy. Lets punctuate this point by revisiting the title of a post 2 years ago by CAIA’s new Associate Director of curriculum, Dr, Keith Black, that really sums up commodity investing: “Commodities not about buy-and-hold.”