« Community Chat, Thurs., Nov. 9, 2006, 7:42 AM | Main | NetEase under the microscope, Thurs., Nov. 9, 2006, 11:22 AM »

November 9, 2006

The re-weighting of commodity indices, Thurs., Nov. 9, 2006, 10:33 AM

The quality and consistency of financial data today is under attack by increasingly skeptical traders. So I thought I'd look into the matter of the Goldman Sachs Commodity Index changes.

There are four major commodity indices used by traders:
1. Dow Jones-AIG (DJ-AIG)
2. Reuters / Jefferies CRB Index (RJ CRB)
3. Rogers International Commodity Index (RICI)
4. Goldman Sachs Commodity Index (GSCI)

These indexes are frequently re-weighted or re-balanced to reflect actual market trading, but there are often huge differences in the weighting methodologies used. Traders, therefore, have to watch whether a major change in the index value is the result of actual price changes in commodity markets or the result of re-weighting of the index components.

The performance of materials/energy linked equities, which had been tracking commodity markets closely, are now, following a re-weighting by Goldman Sachs, showing substantial divergence. The weakness in energy prices from mid-summer has sent the GSCI total return index down roughly 14 pct year-to-date, but the related equity prices don't reflect that, according to a study done by UBS ("Commodities Corner" Oct 31, 2006).

The continuous re-balancing methodology used to create these four major commodity indices is very different. For instance, UBS says that "the target weights of the GSCI are based on world production averages. Hence, whether the price of a commodity has risen or fallen during the year does not cause the GSCI to sell high and buy low as does the DJ-AIG".

UBS also states: "The re-weighting of the Rogers International Commodity Index (RICI) does result in the sale of better performing commodities and vice versa, but since the re-weighting takes place each month, there will be no one-off large impact at the start of 2007. And although the RICI has grown rapidly in 2006, it remains small compared to the GSCI and DJ-AIG".

UBS claims to be unable to estimate the impact of the annual re-weighting of the GSCI commodity index, which is the largest of these four indices (by value invested in financial products that track the indices).

Finally, the UBS Commodity Investment Products Group has stated that with regard to the Reuters / Jefferies CRB Index (RJ CRB): "it is not clear from the Index documentation how often the target weights are set," and so the UBS research department states "consequently there is no way to determine what impact there will be on commodity markets as a result of re-weighting at the start of next year; Financial products benchmarked against the RJ CRB are also much smaller than those benchmarked against the GSCI and DJ CRB".

They conclude:


"Although it is impossible to accurately predict the re-balancing and re-weighting flows of the investment products that track the four major commodity indices we have found that the Dow Jones AIG is almost certainly the most dominant feature of this process due to its design. We cannot calculate the extent of this process as we do not know how much is invested in products that track the index and nor do we know the prices that the exercise will be conducted at: this will only be known in early January 2007. But we have calculated the potential impact of these flows assuming that the prices in January are the same as they were on 19 October.

On this assumption there will be heavy buying of Nat gas, coffee and crude and large selling of nickel zinc, wheat and copper. Substantial selling is likely in nickel and zinc; less selling in copper and aluminum. We suspect these changes will occur during the January roll-period.

We have no information on changes that will be made to the GSCI but these should be small due to a different methodology. The RICI re-weights monthly so there will be no large impact in early 2007. It is not possible to calculate the RJ CRB re-weighting due to a paucity of information but it is likely small.

Accurate predictions are not possible but if prices stay at current levels some large re-weighting and rebalancing flows will occur in January."

These are competent, well-funded researchers speaking. Is it any wonder the rest of us are confused?

Moreover, could the Goldman Sachs re-weighting of the index earlier this year have caused the serious problem for hedge funds like Amaranth, and helped the Administration in the pre-election period?

I don't have an opinion on this, but I'd like to hear from non-GS professionals because I'm sure they do.

Unfortunately, the UBS Commodities Corner report for the 4Q06 is 2 MB (164 pages), which is too large a file for me to upload to my blog publishing software for you. For that reason, I'll be writing short reports like this one, plus some reviews on various commodities, in the next few days.

Posted by Posted by Bill Cara on November 9, 2006 10:33:45 AM | Category: Commodities

Discourse

Barry Ritholtz's Big Picture linked to this sight as the best analysis of the adjustments to the GS index:

http://themessthatgreenspanmade.blogspot.com/2006/10/friends-in-high-places.html

Ritholtz's comments on the article and issue are here:

http://bigpicture.typepad.com/comments/2006/10/blog_spotlight_.html


Posted by: oddlots [TypeKey Profile Page] at November 10, 2006 10:01 PM [link]

Post a comment

Thanks for signing in, . Now you can comment. (sign out)

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)


Remember me?