London, 9 June 2011: Reuters

If you have a nagging suspicion that all is not quite right with commodity prices right now, you could do worse than read the report released over the weekend by UNCTAD.

The less-than-snappy title, "Price Formation in Financialized Commodity Markets", belies a serious attempt to understand how flows of money from the broader investment universe have affected commodity prices and increased the potential for bubbles since the turn of the century.

FINANCIAL SPILL-OVER

And it goes a long way to explaining just why everything seems to be correlated with everything else right now. Once, before the financial crisis of 2008, investment activity in commodities was largely channelled via index funds with their promise of inverse correlation with equities and bonds.

The report suggests that such passive "buy and roll" strategies accounted for 65-85 percent of total investment in the years 2005 through 2008. Index investment brought its own problems, namely the "weight-of-money" impact on cross-commodity prices and the danger that the price effect, compounded by trend-following and trendenhancing system trading, could lead to lengthy periods of disconnect with fundamentals.

Since 2008, however, when everything fell in price together, there has been a growing trend towards actively managed rather than passive investment strategies. The latter's share of investment flows have fallen to just 45 percent in the last couple of years. That brings with it a different type of problem, as the report spells out.

"Financial investors are usually active in several financial markets at the same time. Information collected in one market or for the economy as a whole tends to be used to form expectations about the significant price swings in other markets, regardless of the specifics of supply and demand in the latter.

This mechanism creates new or reinforces existing cross-market linkages, and it increases or alters correlations between two asset classes."

The report drills down further with an illuminating example from last year, namely the release on Dec 10 of the U.S. employment report. It surprised to the downside, triggering a sell-off in several commodities, including oil (WTI) and cocoa.

"It is true that the oil price should have a strong link with economic activity, and that such activity is reflected in labour market data," the report noted. "However, employment and unemployment are lagging indicators, which react rather late in the cycle. Relevant information on activity should therefore already be known from new orders (as an early indicator) or production expectations.

For cocoa, the price reaction cannot at all be explained by fundamentals: there can hardly be any link between United States employment and world chocolate consumption." Such is the manifestation of "the spill-over of the financial market logic to commodity markets". Readers from the world of metals are at this point invited to think of their own recent examples.

INTERVENTION

UNCTAD's conclusions are broadly three-fold: the need for greater information about commodity market fundamentals, more regulatory scrutiny of both on-exchange and off-exchange trading activity and the potential for government intervention in markets.

"As in the case of currency and, more recently, the bond markets, it is possible for a central bank or another agency to engage in the financial markets as a market maker or as the one institution that is able to shock the market when it overshoots."

The report is primarily concerned with food and oil prices rather than metals, which is a shame because the metals markets have been here before and it wasn't exactly a template for government intervention in commodity markets. The International Tin Council (ITC), backed by 23 states and what was then the EC, was a well-intentioned attempt to "regulate" the price of tin using buffer stocks, accumulated during times of low price and released at times of high price.

Unfortunately, the ITC went spectacularly bust in 1985, leading to a cascade of law suits and the near demise of the London Metal Exchange, where the whole sorry saga is still commonly referred to as the "Tin Crisis" for fairly obvious reasons.

The ITC, despite all that sovereign backing, proved no more adept at reading the tin market's fundamentals than any other player in the market. Which makes somewhat suspect the UNCTAD report's assertion that "in markets that are driven by herding, any government agency should be able to understand market developments to the same extent as market participants because it has access to similar information as those participants." Moreover, governmental trading is often even less transparent than private sector trading. Take for example the Chinese government's use of its State Reserves Bureau (SRB) to "regulate" commodity prices, specifically those to which the Chinese economy is most vulnerable.

The SRB has been active in the copper market for many years. Indeed, it has proved to be one of the most influential players. Its confirmation that it was looking to buy in late 2008 effectively put a floor beneath prices. Yet no-one outside of a small handful of Chinese officials has any idea of how much metal it actually holds. It is simply one of the big "known unknowns" in the copper market.

INFORMATION

And, as the UNCTAD report also notes, the number of "unknowns" in the commodities markets is itself a key part of the problem. "Currently, insufficient information makes it difficult for commercial participants to determine whether a specific price signal relates to changes in fundamentals or to financial market events." More than anything else, the report laments the lack of information about stocks.

It concedes that the problem here is that "a significant proportion of stocks is now held privately, which makes information on stocks commercially sensitive." The comments are specifically directed at the agricultural markets but could equally apply to metals, witness the long-running fixation with "hidden stocks" in the copper market. Privately-held stocks have always been "hidden" from the market in the sense that there is no statistically robust way of calculating them.

The problem for metals and for the LME is the increased blurring of "visible" and "hidden" stocks in its own warehouse system. The LME has always, correctly, claimed that exchange-registered inventory and metal availability are not the same thing. After all, someone owns all those stocks at any one time.

What has changed, though, is the encroachment of trading houses into the business of warehousing and stocks financing to the point that the percentage of LME warrants available to the market has dramatically shrunk in the likes of aluminium.

What the LME publishes daily is no longer particularly useful information to market participants. Four million tonnes of registered aluminium, for example, is meaningless if the only available metal is a tiny fraction of that and concentrated in one location and one warehouse operator.

As one of the key commodity exchanges linking the world of financial and "real" commodity markets, the LME could take a leading role in adding to the world's pool of knowledge about commodity stocks. It could for, for example, split its inventory figures between free-float metal and "locked-up" metal, or, as is the practice on both the NYMEX and Shanghai Futures Exchange, provide information of what is not on warrant but still located in LME warehouses.

It would be a small but significant step towards providing the metals market with more accurate understanding of its own dynamics. UNCTAD, meanwhile, might want to suggest to the Chinese government that a bit more transparency in its commodity holdings wouldn't be such a bad thing either.

Ends --

The opinions expressed are his own.