COLUMN-Is the argument from liquidity a fallacy? John Kemp

June 29, 2010

— John Kemp is a Reuters market analyst. The views expressed are his own —

By John Kemp

LONDON, June 29 (Reuters) – What is the “right” level of speculative activity in commodity markets? Different people reach different conclusions, explaining the fierce debate over position limits and other attempts to impose stricter regulation that has broken out since the price spike in 2008.

Economists and market practitioners are divided about the impact of increased participation by investors and speculators over the last decade. Some claim it has improved price discovery and facilitated more hedging. Others blame it for raising volatility, swamping fundamentals and inflating bubbles.

How people answer that first question shapes their response to a second one about how regulators and policymakers should react. Should regulators encourage more participation to improve the market functioning, or should they be trying to restrict it or at least slow it down to safeguard price discovery?

Some economists and practitioners believe more speculation must always be a good thing. If liquidity is good, more liquidity must be better. But that may not be true if the extra speculation is uninformed and simply add “noise” and volatility rather than sharpening price discovery, as leading options expert Paul Wilmott recently argued in his blog (

The argument involves no great point of principle. It is an empirical question about what impact speculation and investment has on price determination, and whether the current level is too much or too little.


There is more common ground than many realise. More or less everyone accepts the primary functions of commodity markets are (1) to facilitate price discovery and (2) to facilitate hedging of physical and perhaps financial risks. Enabling speculators to profit from short and long-term price movements by anticipating them correctly is not a primary function. It is a secondary one that supports the realisation of the primary goals.

The market exists to discover prices and enable producers and consumers to transfer unwanted price risks to counterparties with the opposite exposure, or to speculators who assume the risk in exchange for an (expected) return. If negative hedging pressure (from producers wanting to be short) does not exactly match positive hedging pressure (from consumers wanting to be long) speculators must take up the balance.

Speculators provide both strategic liquidity (when positive and negative hedging demands are not equal) and tactical liquidity (when the timing of hedge sales and hedge purchase does not exactly match).

To the extent speculators simply provide liquidity to one another, rather than to “end users”, there is no net gain and no expected benefit overall. It is simply a gamble, with gains and losses offsetting one another.

The empirical question is how much speculative activity (or “churn”) is needed to provide adequate amounts of strategic and tactical liquidity for producers, consumers and other commercial hedgers to discover prices and offload all the risks they want to shed.

In a very restricted sense, the amount of speculative activity could be quite small (and historically it was). If producers want to hedge 1 billion barrels of crude oil, while consumers want to hedge 800 million, speculators must take up 200 million barrels. Speculators hold only 20 percent of the open interest and appear on only one side of the market.

In practice, speculators hold a range of views about the (uncertain) future so they appear on both sides of the market. The net position must still be 200 million barrels. But gross positions could be several times this number. The question is how big gross positions need to be to provide sufficient liquidity for an efficient net position. How much “excess speculation” does the market need to provide adequate liquidity to the hedgers?


In 1960, Professor Holbrook Working attempted to measure excess speculation in commodity markets in a seminal paper on “Speculation on Hedging Markets”, where he captured it in the form of his famous speculative T index.

“[E]ach commodity appears to have somewhat more speculation than was ‘needed’ to carry the unbalanced short hedging in the commodity. Indeed, the speculative index itself is a direct measure of the amount of that ‘excess’. But at least a large part of what may be called technically an ‘excess’ of speculation is economically necessary.”

There would only be no excess speculation if there were no speculation on the short side, and that could only happen if prices were so low no speculator thought they could go any lower. “A price so low that no speculator thought it likely to go lower would assuredly be too low,” wrote Working.

Reviewing 11 agricultural markets, Working found the amount of excess speculation ranged from 7 percent (wool) to 28 percent (soybeans).

In fact for much of the last 50 years, economists and market participants worried there was not enough speculation, especially at longer-dated maturities, to provide liquidity for all the producers and consumers who might want to hedge longer-term exposure. Even in the most liquid markets, such as crude oil, most liquidity was concentrated in the 1-3 month window, with little or no liquidity for dates beyond 1 year.

The UK Financial Services Authority echoed this viewpoint in its opposition to position limits in December 2009: “We consider that limiting financial participation more generally would hamper market efficiency … Increased participation has brought significant benefits, such as greater depth and liquidity.”

“In particular, we believe greater liquidity should be encouraged in this market, particularly if it facilitates the hedging of the longer maturities that are more closely aligned to the petroleum investment and production cycle. This in turn would enable producers to invest in longer term projects with greater certainty, knowing today what prices they can sell at once the production comes onstream.”


Greater interest in commodity markets from speculators and investors has resulted in a dramatic increase in the number of open contracts (and by implication liquidity) since 2004, especially in longer-dated futures and options, where swap dealers, hedge funds and other non-commercials dominate the market according to CFTC research (Charts 1-2).





Most liquidity is still restricted to the first 24-36 months, however, and the increase in liquidity beyond 2 years seems to have stalled since 2008 (Chart 3).

Moreover, it seems that speculative activity is outpacing the growth the hedging interest. Calculations of the speculative T index using the CFTC’s disaggregated data show the amount of “excess” speculation in WTI-linked oil futures and options rose from 20-40 percent to peak at 70-90 percent at the height of the 2008 oil crisis, before gradually subsiding gradually below 60 percent in the last year (Chart 4).

Given what we know about the concentration of hedging in the nearby maturities and dominance of non-commercial positions at longer-dates, the amount of excess speculation in the far forward market is much higher.

The question of whether all this extra liquidity is needed depends on whether there is a substantial volume of “latent hedging” producers and consumers would like to do, but which has been frustrated in the past by a lack of liquidity.

Both dealers and regulators need to develop a clearer understanding of just how many producers, consumers and perhaps financial institutions want to “hedge” their exposure, beyond 1-2 years forward. At the moment the scale of the demand is unclear. Fixing prices far into the future is expensive (via options) and carries significant opportunity cost (via futures).

It is far from clear if there really is much latent hedging from producers who would like to lock in prices for the next 5-10 years, or consumers who want to protect themselves against price rises over a decade. If there is, regulators should encourage more financial participation, not less. If there isn’t, more speculative activity is simply unproductive churn. ((; Reuters messaging:; +44 207 542 9726))

No comments so far

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see