Conquering Misperceptions About Commodity Futures Investing

Duke University – Fuqua School of Business; National Bureau of Economic Research (NBER)

August 20, 2015

Abstract:

Three fundamental misperceptions plague many commodity futures investors and have probably contributed to a sense that commodity futures investments have not lived up to expectations. The first misperception is that commodity futures are a play on commodity prices. The futures return includes both a price return and an “income return” (from rolling into the next futures contract as well as a collateral return). In The Strategic and Tactical Value of Commodity Futures, Erb and Harvey (2005) show the income return is by far the most important component of the futures total return. This income return could be positive (as it was in the past) or negative (as it has been in the recent past and may be in the future). The second misperception is that commodity futures are an asset class. Erb and Harvey argued that portfolios of commodity futures are investment strategies, not an asset class, and consistent with the first misperception, portfolios of commodity futures are more a play on “income returns” than a play on price returns. Expecting commodity futures investments to offer the pay-off of an asset class driven by commodity prices is a recipe for frustration. Finally, there is a misperception that commodity futures returns deliver “equity like” performance. Erb and Harvey argued against the presumption that a passive investment in commodity futures would produce “equity-like” returns in the longer term. They argued that the real price return to commodities may be negative over the long run mainly due to technological change. This should lead to modest expectations of future price appreciation. More importantly, the income return is the main driver of total returns (and could be positive or negative). We present out of sample evidence from 2004-2015. A passive investment in commodities futures has produced negative average returns in this period and almost all of that negative performance is driven by the income return – not by a decline in commodity spot prices.

Related research:

The Strategic and Tactical Value of Commodity Futures

The Golden Dilemma

The Golden Constant

An Impressionistic View of the ‘Real’ Price of Gold Around the World

Conquering Misperceptions About Commodity Futures Investing – Introduction

Investing in commodity futures is an enigma to many. It is widely believed that if the price of a commodity goes up, it must be the case that an investment in the futures would be profitable. This is not necessarily the case. Indeed, there are three components to a commodity futures investment return: the “price return” (for the leading commodity futures indices, the price return is the percentage change in the “spot” price), the roll return (profit or loss you make from rolling into a new contract) and the collateral return. The roll return combined with the collateral return can be referred to as the “income return” on a commodity futures investment.

In The Tactical and Strategic Value of Commodity Futures, Erb and Harvey (2005) show that historically long?term return differences across individual commodity futures were largely driven by long?term income returns, not long?term price returns. This update suggests that the long?term total returns of diversified commodity futures portfolios may also largely be driven by long?term income returns, not long?term price returns. This perspective helps explain the disappointing returns that some investors might have experienced with some diversified commodity futures portfolios over the last decade.

Rather than thinking about portfolios of commodity futures as bets on commodity prices they might more fruitfully be viewed as bets on income returns ?? returns that are driven by the term structure of futures prices. This can be problematic for a few reasons. First, many investors may have little or no idea that individual commodity futures have a term structures of prices, what a term structure happens to be at any point in time or how term structure influences affect observed total returns. Second, those investors who are unfamiliar with commodity futures term structures may have little inclination to learn about commodity futures term structures. Finally, investors seeking to exploit possible insights into the overvaluation or undervaluation of commodity prices may find their valuation insights swamped by the income return impact of commodity term structures.

Revisiting the Track Record

Exhibit 1 decomposes the total return of a commonly used commodity futures index, the S&P GSCI, into the key return drivers: income and a price return (as is the case for most other assets). The S&P GSCI is used purely for illustrative purposes. This is not meant to be a post?mortem of the performance of the S&P GSCI. Rather the goal is to highlight issues that very likely apply to many commodity futures portfolios. Erb and Harvey (2005) and Gorton and Rouwenhorst (2005) examined the performance of individual commodity futures and portfolios of commodity futures through 2004. The date of these studies provides a rationale for examining commodity futures returns up to 2004 and since 2004. The reason to focus on a commodity futures index such as the S&P GSCI is that it has historically been the most widely used commodity index. The goal of this update is to observe what worked or did not work with “commodity investing” since 2004, not to provide an encyclopedic overview of the performance of the many different commodity futures indices that now exist.

Exhibit 1 shows that through 2004 the S&P GSCI had an income return of 8.7%, a price return of about 3.2% per year and an 11.9% total return. Seen from the vantage point of December 2004, this can be viewed as an in?sample return, the known historical performance of the S&P GSCI. Interestingly the income return accounted for about 73% of the total return and the price return.