Thursday, December 19, 2013

Self-Indexing in Commodity-Linked Investments - Citi CUBES

By Tim Dulaney, PhD, FRM and Tim Husson, PhD

Over the past few weeks and months, we've noticed a pattern in the products coming across our desks: structured investments linked to esoteric proprietary indexes, created by the same bank that issued the product.  We touched on this topic a bit when we discussed self-indexing in the context of structured certificates of deposit, but we thought we'd revisit the issue with a few of the examples that we've been looking into more recently.  The examples we'll discuss each reference a proprietary strategy for obtaining exposure to commodity futures markets.1

In September 2009, Citi launched a set of indices that tracked exposure to the commodity futures market, dubbed the Citi Commodities Curve Beta Enhanced Indices or CUBES.

Citi's CUBES operate under the assumption that the term structure of a given commodity is "expected to remain fairly constant" and that "any significant deviation from this relationship is unsustainable in the short term."  At the end of each month, Citi averages observed futures contract prices for the current month and the previous month.  This average is then used as the estimate for the term structure for the end of the following month.  Citi then selects the contract that will have the largest expected yield in one month.

This strategy will typically result in exposure to longer-dated futures contracts (futures contracts that are further out in the term structure) where prices do not vary as much. This strategy attempts to limit the negative effects of roll-yield in contangoed commodity markets.2 More information can be found in the SEC filings for the indexes.

Like most newly created indexes, the strategy backtests well.  In particular, Citi mentioned that the "S&P GSCI weighted Citi CUBES excess return aggregate index has generated returns of 14.60% [...] with an annualized volatility of 21.84%" when the S&P GSCI index has returned 4.37% with a volatility of 25.37% for the period 1999-2009.

It is important to note that Citi's CUBES methodology depends a great deal on its assumption about commodity term structures remaining constant.  The academic literature on the term structure of futures markets is voluminous, and the precise dynamics of these markets is still not completely understood.  Investors should appreciate that any investment linked to Citi's CUBES indexes represents a bet on Citi's potentially simplistic view of commodities markets.

Since the launch of the indexes, Citi has created funds that track these indices.  Having a fund or product that tracks an index that the sponsor/issuer themselves is responsible for calculating engenders deep conflicts of interest.  When, for example, the payment on debt is linked to the performance of one of these indexes, there is an incentive for the calculation agent to alter the performance of the index to decrease the bank's liabilities, reducing returns to investors.

Issuers may argue that by developing their own indexes they are reducing the fees that they would otherwise have to pay to license a third party's index, but in fact using a third party index helps reduce conflicts of interest.  As more and more exchange-traded products, structured products, structured CDs, and other investments move to self-indexing, these conflicts may become ever more common.  We will continue to track this issue closely.
A futures contract is a standardized contract that represents an exchange of money between two parties for the obligation to deliver a very specific commodity at some date in the future.  The prices of futures contracts on the same commodity across different delivery dates forms a term structure.
For more information about the effects of roll-yield on the profitability of futures-based strategies, see our blog posts and research on the topic.

No comments:

Post a Comment

Please keep comments appropriate. Malicious comments or solicitations will be removed.