Two major variables are predictors of commodity index returns:commodity weights and futures contract selection.
The major commodities indexes—the Deutsche Bank Liquid Commodities Index (DBLCI), Dow Jones-UBS Commodities Index (DJCI), S'P GSCI Index (GSCI), Thomson Reuters Equal Weight Continuous Commodities Index (CCI), SummerHaven Dynamic Commodity Index (SDCI) and the Rogers International Commodity Index (RICI)—all have wildly different portfolios.
While these differences are stark and obviously play a significant role in determining returns, the method of choosing and rolling futures contracts can play as big, or even a bigger, role in driving returns—especially in energy-heavy indexes like the GSCI that tend to have a large percentage of their portfolio in commodities that are in contango.
When a commodity is in contango, the price of the current futures contract is less than the price of the next-month futures contract, meaning you lose money when your exposure is rolled from an expiring current futures contract into the one that expires in the following month.
Generally, the cost of rolling exposure varies across different contract maturities.
Soybean futures contracts, for example, are in contango in the near term, but in backwardation further down the curve. Backwardation, the opposite of contango, occurs when the contract that you roll into is cheaper than the contract that you currently hold, allowing you to pick up a bit extra profit when the exposure rolls.
A commodities index that limited its exposure-rolling system to the next-month contracts would lose money at roll time, while one that optimized its rolling system to minimize the effects of contango, or, where applicable, maximize backwardation, would profit. Which brings me to my point …
Why is anybody still investing in front-month commodities ETFs? I ask, because they are, and in a big way.
Two of the most popular commodities exchange-traded products offer front-month exposure. They are the iPath Dow Jones-UBS Commodity Total Return ETN (NYSEArca:DJP - News) and the iShares S'P GSCI Commodity ETF (NYSEArca:GSG - News), and they have $2.46 billion and $1.44 billion in assets, respectively.
Of course, the DJ-UBS and GSCI indexes are long-standing, well-established funds, so I understand why investors are attached to them.
They each offer unique portfolios—DJ-UBS uses a proprietary rules-based methodology to select and weight its commodities, while the GSCI chooses its commodities based on the liquidity of their futures contracts and weights them based on the production of each commodity.
But what if you could keep the commodities selected by your chosen index, in the weights specified by the index provider, while also optimizing the selection of futures contracts to maximize roll yield?
You can, but the idea hasn’t really caught on.
The next-generation iPath Pure Beta S'P GSCI-Weighted ETN (NYSEArca:SBV - News) does just that with the same S'P GSCI Index, and yet, surprisingly enough, it hasn’t picked up many assets since its launch in April 2011. Its $5 million in assets really looks puny next to GSG’s $1.44 billion.
Deutsche Bank has a series of commodities indexes—dubbed “Commodities Boosters”—that replicate the selection and weighting of the DJ-UBS Commodities Index and the S'P GSCI, as well as subindexes of those two broad indexes that focus on agriculture and energy, among others.
The Deutsche Bank Commodity Booster linked to the DJ-UBS index outperformed the standard DJ-UBS index by nearly 15 percent over the past five years, seemingly due to futures contract selection and roll-timing .
Similarly, the Deutsche Bank Commodity Booster based on the S'P GSCI outperformed the standard GSCI by 26 percent over the same time period.
Roll optimization gets more important as commodities with high weights within an index go into steeper contango.
As you can see from the two graphs above, the roll-optimized ETFs really pulled away from the traditional indexes in late 2008/early 2009—right when crude oil moved into steep contango. Crude oil currently makes up nearly half of the GSCI, so it’s not surprising that the difference is more pronounced in the GSCI.
That said, the Deutsche Bank Commodity Booster indexes are currently only tracked by European-listed ETFs. Here’s to hoping they make it over to U.S. exchanges sometime soon.
Until then, the PowerShares DB Commodity Index Tracking Fund (NYSEArca:DBC - News) will do just fine. DBC uses the same optimum yield futures contract selection process, albeit on a different set of commodities than the GSCI and DJ-UBS indexes mentioned above.
DBC consistently outperforms competitors like GSG and DJP -- over the past three years, it has returned 51.2 percent, compared to GSG’s 45.1 percent and DJP’s 35.8 percent.
iPath’s suite of Pure Beta ETNs are also good options for mitigating the effects of contango or taking advantage of commodities in backwardation.
My point is this:If you’re still investing in a front-month rolling ETF, there are better options out there, and I encourage you to explore them.
Permalink | ' Copyright 2012 IndexUniverse LLC. All rights reserved
More From IndexUniverse.com