By Scott Irwin
Econbrowser is pleased to host another contribution from Scott Irwin, who holds the Laurence J. Norton Chair of Agricultural Marketing at the University of Illinois, and today offers some insights from his research on the current debate concerning commodity speculation.
Index Funds and Commodity Prices… Here We Go Again
by Scott Irwin
by Scott Irwin
Some issues just will not go quietly into the night. The U.S. Senate’s Permanent Subcommittee on Investigations released a report on June 23, 2009 concluding that excessive speculation by large index funds in the Chicago Board of Trade (CBOT) wheat futures market resulted in over-valued futures prices, a large “carry” in the futures price structure, and a wide divergence between futures and cash prices during the 2006 through 2008 period. These are serious charges, and judging by recent newspaper headlines, there is a receptive audience in the halls of Congress and on the part of some regulators.
If you read my previous Econbrowser posts on commodity markets ( here and here ) you should not be surprised to discover that I disagree with Subcommittee’s conclusions. Rather than citing rigorous academic analysis of the problems in the CBOT wheat contract the Subcommittee chose to rely on finger-pointing by industry participants who were adversely affected by high prices and weak basis and on cursory analysis of temporal price, basis, and trading activity data. You can find an extended commentary on the Subcommittee report here by our research group at the University of Illinois.
To begin, the Subcommittee repeats the mistake of other “bubble” proponents by equating index fund money flows into wheat futures with demand for the physical commodity. Simply observing that large investment dollars flowed into the long side of the wheat futures market at the same time that futures prices rose substantially does not necessarily prove anything. Several recent studies use “Granger causality tests” and “cross-sectional” regression tests to establish that lagged position changes by any group, including index traders, do not help to forecast current futures price changes in a statistically significant manner (see our commentary and this paper for a brief review of the studies). This provides compelling evidence that index funds were not responsible for the run-up in grain and oilseed markets, particularly in the wheat market. Historically, price “spikes” have not been uncommon in the grain markets as market participants react to developments that are less permanent than anticipated. Such episodes are frequently attributed to speculation rather than the underlying fundamental factors responsible for the market situation.
If problems in the CBOT wheat futures market were not due to a speculative bubble in futures prices, then what was the problem? Our research at the University of Illinois pinpoints two major factors contributing to the lack of convergence between cash and futures price in wheat. The first factor is the tendency for spreads in the futures market to reflect a relatively high percent of full carry (contango) since 2006. The second factor is long-term structural deficiencies in the delivery system for CBOT wheat.
Large carrying charge markets contribute to lack of convergence by “uncoupling” cash and futures markets when futures prices are above cash prices. The delivery instrument for wheat was a warehouse receipt until recently when it was changed to a shipping certificate (starting with the July 2008 contract). Those longs who receive certificates or receipts from shorts in the delivery process are not required to cancel those instruments for shipment. The instruments can be held indefinitely with the holder paying "storage" costs at the official rates specified by the CBOT in contract rules. The taker in delivery (the long) may choose to hold the delivery instrument rather than cancelling the instrument by load out of physical grain if the spread between the price of the expiring and next-to-expire futures contracts exceeds the cost of owning the delivery instrument. Therefore, as the magnitude of the nearby spread exceeds the full cost of carry for market participants with access to low-cost capital, those participants can (and do) stand for delivery but do not cancel delivery certificates or receipts for load out. The lack of load out, then, means that deliveries do not result in cash commodity purchases by the taker that would contribute towards higher cash prices and better convergence.
The chart below illustrates the relationship between the magnitude of the carry in wheat and the basis (cash – futures price) at the primary delivery location for wheat, Toledo , on the first delivery date of each expiring wheat futures contract over March 2000 through July 2009. The percent of full carry is computed as (price of next-to-expire contract minus price of expiring contract)/(storage plus interest costs).
The chart is constructed so that the zero line for basis on the left y -axis scale corresponds to 80% of full carry on the right y -axis scale. While the pattern is not perfect, there is a consistent tendency towards poor convergence (wide basis) whenever the carry exceeds about 80% and better convergence when the carry is below 80%. Note that this pattern is evident not only during recent years but also in 2000-2001, when wheat experienced another period of non-convergence, albeit at much smaller basis levels. A similar relationship exists for corn and soybeans.
Of course, this only moves the debate back one step to explaining the large carrying charge in the wheat futures contract since 2006. The Senate Subcommittee concludes that the rolling of positions by index funds is responsible for the persistence of the large carry. Rolling refers to in dex funds entering market positions in the nearby contract and then moving the positions to the next contract before the maturity of the nearby contract. The evidence, however, does not support a conclusion that such rolling widened the spreads.
The following chart shows the behavior of nearby spreads for CBOT wheat during the first 13 business days of the calendar month prior to contract expiration for the March 1995 through March 2009 contracts. The time window for the analysis is centered on days 5 through 9, the time period of the so-called “Goldman roll” when index funds tend to roll their positions from the nearby to the next deferred contract.
The averages reveal a consistent increase in the size of the spread to the next contract (expressed as a percent of full carry) during “Goldman roll” days 5 through 9. However, the spike in the magnitude of the spread either disappears entirely or noticeably recedes during days 10 through 13, so rolling did not necessarily lead to a permanent increase in the magnitude of the spread. The spike in the magnitude of the spread during the roll period was also present long before convergence became an issue and before index funds had a major presence in these markets. This is not surprising since the time window when index funds roll to the next contract is also the same time period when many other traders roll their positions.
As noted above, the second factor contributing to recent convergence problems is underlying structural problems with the CBOT wheat delivery system. The fundamental problem is that changes in wheat production patterns, transportation logistics, and trade flows have left the contract with an increasingly narrow commercial flow of wheat to draw upon in the delivery process. For example, annual commercial shipments of wheat at facilities regular for CBOT delivery averaged only 28 million bushels over 2000-2008. By comparison, annual commercial shipments for corn and soybeans at delivery facilities averaged of 260 and 170 million bushels, respectively, over the same time period. This raises substantial doubts about the representativeness of Chicago and Toledo as wheat pricing and delivery points. Furthermore, there is a constant potential for congestion in the delivery process of CBOT wheat futures and the attendant distortion of cash and futures prices.
The underlying structural problems were generally ignored until recently when additional delivery locations were added by the CBOT starting with the July 2009 contract. Whether these additional wheat delivery locations will contribute to improved convergence performance is debatable due to the “safety valve” nature of the new location price differentials. Early evidence from the expiration of the July 2009 CBOT wheat contract is not encouraging.
Fundamental questions still remain regarding the performance of the CBOT wheat futures contract. First, what explains the increase in the level of carrying charges in the wheat futures markets since 2006? I believe it is likely a combination of CBOT contract storage rates that lagged market rates, congestion in the delivery process, and an increase in risk that had a large impact on stockholding behavior. Second, why does t he CBOT wheat contract, the most popular wheat contract in the world, appear to be widely used to trade “wheat” generically when the delivery market locations make the contract a soft red winter wheat contract at maturity? To the extent that world wheat and soft red winter wheat supply and demand fundamentals diverge, this can lead to a systematic tendency towards poor convergence performance. Third , what explains the seemingly anomalous behavior of commercial grain firms (e.g. ADM , Bunge , and Cargill ) during recent episodes of non-convergence? These firms are “regular” for delivery and can create delivery certificates virtually at will. As Craig Pirrong noted recently, it is not clear why these firms have apparently left so much money on the table by not arbitraging the very large differences between cash and futures prices. Understanding the motivation and strategies of these firms would be a far more informative line of inquiry than the current obsession with index funds.