A partnership with academia

Building knowledge for trade and development

Vi Digital Library - Text Preview

Price Formation in Financialized Commodity Markets - The Role of Information

Report by UNCTAD, 2011

Download original document (English)

This study asserts that the role of information is crucial for price developments in commodity derivative markets. It demonstrates that the traditional approach of efficient market hypothesis (EMH)does not apply for present commodity futures markets. Trading decisions are taken in an environment of considerable uncertainty, where traders engage in "intentional herding" and do not base their transactions on information about fundamentals. This analysis also advocates several policy responses to improve market functioning such as increased transparency with respect to fundamentals, increased transparency in the exchanges and OTC markets, tighter regulation of financial players and the introduction of a transaction tax system.

U N I T E D N A T I O N S C O N F E R E N C E O N T R A D E A N D D E V E L O P M E N T
A N D A R b E I T E R k A M M E R W I E N


Price formation in
financialized commodity markets


The role of information


EMBARG
O


The conte
nts of this


Publicatio
n


must not b
e quoted


summariz
ed in the p


rint, broad
cast


or electron
ic media b


efore


5 June 2
011, 17:0


0 hours G
MT






Price Formation in Financialized
commodity markets:


the role oF inFormation


Study prepared by the secretariat of the
United Nations Conference on Trade and Development


UNITED NATIONS
New York and Geneva, June 2011




ii


Note


Symbols of United Nations documents are
composed of capital letters combined with
figures. Mention of such a symbol indicates a
reference to a United Nations document.


*


* *


The designations employed and the presentation
of the material in this publication do not imply
the expression of any opinion whatsoever on
the part of the Secretariat of the United Nations
concerning the legal status of any country,
territory, city or area, or of its authorities, or
concerning the delimitation of its frontiers or
boundaries.


*


* *


Material in this publication may be freely
quoted or reprinted, but acknowl edgement
is requested, together with a reference to the
document number. A copy of the publication
containing the quotation or reprint should be
sent to the Publications Assistant, Division on
Globalization and Development Strategies,
UNCTAD, Palais des Nations, CH-1211
Geneva 10.


UNITED NATIONS PUBLICATION


UNCTAD/GDS/2011/1


Copyright © United Nations, 2011
All rights reserved




iiiPrice Formation in Financialized Commodity Markets: The Role of Information


Table of contents


Page


Abbreviations ...............................................................................................................................................v
Acknowledgements ......................................................................................................................................vi
Executive summary ....................................................................................................................................vii


1. Motivation of this study .......................................................................................................1


2. Price forMation in coMModity Markets ....................................................................3
2.1. Information and commodity price formation ..................................................................................3
2.2. The role of futures exchanges and OTC markets in commodity price formation ..........................6


3. recent evolution of Prices and fundaMentals .....................................................9
3.1. Crude oil ..........................................................................................................................................9
3.2. Selected food commodities ............................................................................................................10


4. financialization of coMModity Price forMation ..............................................13
4.1. Financialization:definition,motivation,sizeandinstruments ......................................................13
4.2. Categories of market participants ..................................................................................................18
4.3.Whatisproblematicaboutfinancialization? ................................................................................20
4.4. Herdbehaviourandthelimitsofarbitrage ....................................................................................20
4.5. Thepriceeffectsofthefinancializationofcommoditymarkets ..................................................24
4.6. Herding and its effects in different markets ..................................................................................29
4.7. Commodity prices and world business cycles ...............................................................................34


5. field survey .................................................................................................................................37
5.1. Objectives ......................................................................................................................................37
5.2. Choice of participants ....................................................................................................................37
5.3. Approach ......................................................................................................................................38
5.4. Results ..........................................................................................................................................41
5.4.1. Physical traders ...................................................................................................................41
5.4.2. Financial players .................................................................................................................45
5.4.3. Others .................................................................................................................................47
5.5. Summary ........................................................................................................................................48


6. Policy considerations and recoMMendations ....................................................49
6.1. Improvingtransparencyinphysicalcommoditymarkets..............................................................50
6.2. ImprovingtransparencyinfuturesandOTCcommoditymarkets ...............................................51
6.3. Tighterregulationoffinancialinvestors ........................................................................................52
6.4. Price stabilization schemes and other mechanisms .......................................................................53


7. conclusions ..................................................................................................................................55


annex ........................................................................................................................................................57
notes ..........................................................................................................................................................59
references .................................................................................................................................................63
Glossary ....................................................................................................................................................67




iv


List of tables


1 Leading exchanges for oil and agricultural commodity derivatives ..................................................8
2 Trader categories in the CFTC’s Disaggregated Commitment of Traders reports ..........................19
3 Simultaneous correlation between price and position changes, selected commodities


and trader categories, July 2009–February 2011 .............................................................................28
4 Classification of interviewees .........................................................................................................37
A.1 World production of biofuels ...........................................................................................................57


List of figures


1 Evolution of crude oil prices, 1980–2010 .........................................................................................9
2 Evolution of grain prices, 1980–2010 .............................................................................................10
3 Evolution of prices of selected soft commodities, 1980–2010 ........................................................12
4 Futures and options contracts outstanding on commodity exchanges,


December 1993–December 2010 ....................................................................................................15
5 Notional amount of outstanding over-the-counter commodity derivatives,


December 1998–June 2010 .............................................................................................................15
6 Financial investments in commodities: assets under management, by product, 2005–2011 ..........16
7 Financial investments in commodities and global GDP, 1998–2010 ..............................................17
8 Financial investments in commodities and global oil production, 2001–2010 ...............................17
9 Different types of herd behaviour ....................................................................................................21
10 Actual price developments and estimated price developments without index investors,


selected commodities, 2006–2009 ...................................................................................................26
11 Actual and predicted crude oil prices, 1997–2008 ..........................................................................27
12 Maize: prices and net long financial positions, by trader category, June 2006–February 2011 .........27
13 Crude oil: prices and net long financial positions, by trader category, June 2006–February 2011 ....28
14 Thirty-day rolling correlation between the WTI front month futures contract and


the Australian dollar–United States dollar exchange rate, 1986–2010 ............................................30
15 Thirty-day rolling correlation between the WTI front month futures contract and


the S&P 500, 1986–2010 .................................................................................................................30
16 Thirty-day rolling correlation between the DJ-UBS Agriculture Total Return Index and


the United States dollar–Brazilian real exchange rate, 1992–2010 .................................................31
17 Thirty-day rolling correlation between the WTI front month futures contract and


the United States dollar–Brazilian real exchange rate, 1992–2010 .................................................31
18 Relationship between the Brazilian real–Japanese yen exchange rate and selected commodity


markets, August 2008–July 2010 .....................................................................................................32
19 Effects of announcement of employment data in the United States (rebased series),


3 December 2010 .............................................................................................................................33
20 Dynamics of world industrial production after the peaks of four business cycles ..........................34
21 Commodity prices and market indexes before and after the trough of September 1980 ................35
22 Commodity prices and market indexes before and after the trough of December 1982 .................35
23 Commodity prices and market indexes before and after the trough of December 2001 .................35
24 Commodity prices and market indexes before and after the trough of February 2009 ...................35
A.1 Prices and net long financial positions, by trader category, selected commodities,


June 2006–February 2011 ................................................................................................................58




vPrice Formation in Financialized Commodity Markets: The Role of Information


Abbreviations


CBOT Chicago Board of Trade
CFTC Commodity Futures Trading Commission
CIT commodity index traders
COT commitments of traders
CPB Central Planning Bureau Netherlands Bureau for Economic Policy Analysis
DCOT disaggregated commitments of traders
DJ-UBSCI Dow-Jones-Union Bank of Switzerland Commodity Index
ECB European Central Bank
EIA Energy Information Agency
EMH efficient market hypothesis
ETF exchange-traded fund
ETN exchange-traded note
ETP exchange-traded product
EU European Union
FAO Food and Agriculture Organization of the United Nations
HFT high-frequency trading
ICE Intercontinental Exchange
IEA International Energy Agency
IMF International Monetary Fund
IOSCO International Organization of Securities Commissions
ISO International Sugar Organization
JODI Joint Organisations Data Initiative
LIFFE London International Financial Futures Exchange
NYMEX New York Mercantile Exchange
OECD Organisation for Economic Co-operation and Development
OPEC Organization of the Petroleum Exporting Countries
OTC over the counter
PMPU producers, merchants, processors, users
S&P GSCI Standard & Poor’s Goldman Sachs Commodity Index
UNCTAD United Nations Conference on Trade and Development
USDA United States Department of Agriculture
WTI West Texas Intermediate




vi


Acknowledgements


This study was prepared by the UNCTAD secretariat for Arbeiterkammer Wien (Austria). It was prepared
by a research team consisting of Heiner Flassbeck (team leader), Director, Division on Globalization and
Development Strategies, David Bicchetti, Jörg Mayer and Katja Rietzler (independent consultant). Pilar
Fajarnes and Nicolas Maystre provided specific inputs. Makameh Bahrami helped with the data. The study
was edited by Praveen Bhalla.


The financial support of Arbeiterkammer Wien is gratefully acknowledged.




viiPrice Formation in Financialized Commodity Markets: The Role of Information


executive summary


The mid-2000s marked the start of a trend of steeply rising commodity prices, accompanied by increasing
volatility. The prices of a wide range of commodities reached historic highs in nominal terms in 2008 before
falling sharply in the wake of the financial and economic crisis. Since mid-2009, and especially since the
summer of 2010, global commodity prices have been rising again. These developments coincide with major
shifts in commodity market fundamentals, particularly in emerging economies which are experiencing fast
growth, increasing urbanization and a growing middle class with changing dietary habits, including an
increasing appetite for meat and dairy products. In addition, in an attempt to reduce the use of fossil fuels in
energy consumption, a range of food crops are now being used in the production of biofuels, which is being
promoted in a number of countries including those of the European Union (EU) as well as the United States.
The related conversion of land use from crops for food to crops for biofuel production has also affected the
prices of food crops. At the same time, a decline in the growth rates of production and productivity, partly due
to the adverse effects of climate change, has adversely affected the supply of agricultural commodities.


However, these factors alone are not sufficient to explain recent commodity price developments; another
major factor is the financialization of commodity markets. Its importance has increased significantly since
about 2004, as reflected in rising volumes of financial investments in commodity derivatives markets – both
at exchanges and over the counter (OTC). This phenomenon is a serious concern, because the activities of
financial participants tend to drive commodity prices away from levels justified by market fundamentals,
with negative effects both on producers and consumers.


The role of information flows is crucial for price developments in commodity derivatives markets.
Traditionally, the so-called efficient market hypothesis (EMH) is assumed to hold in financial markets,
including in commodity derivatives markets and especially in futures markets, which are the focus of
this study. The EMH postulates that all publicly available information is immediately reflected in prices.
In its strong form, the EMH contends that even private information – available only to individual market
participants – is reflected in the price through the effects of the transactions of the persons in possession
of the information. If the EMH were to apply, commodity price developments would reflect nothing but
information on fundamentals.1


However, this study shows that the EMH does not apply to the present commodity futures markets. Market
participants also make trading decisions based on factors that are totally unrelated to the respective commodity,
such as portfolio considerations, or they may be following a trend. Therefore, it is difficult for other agents
in the market to discern whether or not their transactions are based on information about fundamentals,
which in any case is sometimes difficult to obtain and not always reliable. Trading decisions are thus taken
in an environment of considerable uncertainty. In such a situation, it is rational to follow other participants’
trading decisions. A wide range of motivations leads traders to engage in this so-called “intentional herding”
on a perfectly rational basis, the most important one being imitation in situations where traders believe that
they can glean market information by observing the behaviour of other agents.


In an environment of herd behaviour there are limits to arbitrage. Acting against the majority, even if justified
by fundamentals, may result in large losses, often of borrowed money. It may therefore be rational for market
participants to ignore their own information and follow the trend. This is what many financial players do by
default, basing their trading decisions purely on the behaviour of price series (algorithmic trading), which
can lead to a commodity price bubble.




viii


There is considerable empirical evidence that points to financial investors’ impact on commodity prices:


A number of studies find evidence of commodity price bubbles. Analyses show that position-taking •
by index investors, that passively replicate the price movements of an index based on a basket of
commodities, has an impact on price developments, particularly of crude oil and maize. The fact
that these effects are persistent – especially in the case of crude oil – points to the presence of herd
behaviour. Whereas index investors were identified as significant price drivers prior to the financial
crisis, the importance of money managers (e.g. hedge funds), that follow more active trading strategies
and take positions on both sides of the market, has increased since then. This is reflected in the very
close correlation between price changes and position changes of money managers since 2009, which
is as high as 0.8 in the oil market. Indeed, it has been estimated that speculation currently accounts for
as much as 20 per cent of the oil price.


Cross-market correlations between currency and commodity markets have increased recently, and •
point to factors other than fundamentals that are driving commodity prices. Information flows in other
financial markets increasingly influence the dynamics of commodity futures. In addition, an analysis of
the reactions of commodity prices to announcements of economic indicators shows that, within minutes
of an announcement, commodity prices react in a similar manner across different commodity markets
that do not have much in common.


The behaviour of commodity prices, especially oil, over the business cycle has changed fundamentally. In •
earlier business cycles commodity prices and equity prices evolved differently. Increases in commodity
prices did not occur until well after the trough. In the most recent business cycle, on the other hand,
oil prices surged immediately after the trough, even before share prices started to rise. This surge was
based simply on the expectation, not the actual occurrence, of an upswing.


To complement the theoretical and empirical findings 22 interviews were conducted with various commodity
market participants, ranging from physical traders to financial investors, but also including a broker,
representatives of a price reporting firm and two consultants. The interviewees agreed that the role of financial
investors has become more important in recent years. Due to their financial strength, they can move prices
in the short term. This leads to increased volatility, which may harm markets and drive hedgers with an
interest in physical commodities away from commodity derivatives markets. The increased volatility results
in more margin calls and thus higher financing requirements. Although all interviewees stressed the role
of fundamentals in medium- to long-term commodity price formation, they conceded that substantial price
distortions and herding effects could occur in the short term due to the participation of financial investors.
This is also reflected in the responses of several interviewees, who said they paid increasing attention to
financial market information. The main conclusion of the interviewed commodity market players was that
market transparency needed to be increased. For the United States, this refers especially to the OTC market.
In Europe, there is, in general, a greater lack of transparency than in the United States. The adoption of
reporting in Europe, similar to that provided by the Commodity Futures Trading Commission (CFTC) – the
institution mandated to regulate and oversee commodity futures trading in the United States – in its weekly
Commitments of Traders reports would be a big step in the right direction, but more information should also
be required about the OTC business. Concerning other regulatory issues, the level of awareness of current
discussions on regulation and reform differed widely among the interviewees. Generally, they appeared to
have paid more attention to United States regulations, such as the Dodd-Frank Act, whereas only a minority
of those interviewed had a clear idea about the European Commission’s regulatory initiatives. There was
substantial scepticism about bans (e.g. on high-frequency trading) and position limits. The general belief
was that regulations were rather difficult to enforce.


The analysis clearly shows that information flows play a vital role in commodity price developments. The
market distortions described above are closely related to the fact that market participants make decisions
under conditions of substantial uncertainty. Therefore policy responses to improve market functioning should
concentrate on the following issues:




ixPrice Formation in Financialized Commodity Markets: The Role of Information


Increased transparency with respect to fundamentals. Although a variety of sources of information •
currently exist, there is substantial uncertainty in terms of data quality and timeliness, particularly with
respect to inventories.


Increased transparency in the exchanges and OTC markets themselves. More information should be •
made available with regard to position-taking and categories of market participants in commodity
derivatives markets. This applies in particular to commodity trading in Europe, where transparency
lags significantly behind that in United States exchanges. Improved transparency is important not only
for market participants but also for regulators, who can only intervene if they know what is happening
in the market.


Tighter regulation of financial players. Tighter rules internationally would be an optimal scenario, so •
that regulatory migration could be avoided. Given that the size of financial players’ involvement has a
substantial impact on price developments, position limits aimed at restraining the engagement of financial
investors in commodity markets may be indispensable in the medium to long run. As appropriate levels
are not easy to determine, a first step might consist of position points at which traders would be required
to provide additional information. In addition, proprietary trading by financial institutions that are
involved in hedging transactions of their clients could be prohibited because of conflicts of interest.


Beyond this kind of “soft regulation”, a number of direct commodity price stabilization measures •
should be considered. As governments and international institutions have access to the same kind of
information as the market participants, the establishment of a government-administered virtual reserve
mechanism and the possibility of allowing governments’ direct intervention in the physical and the
financial markets need to be considered. In financialized commodity markets, as in currency markets,
intervention may even help market participants to better recognize the fundamentals.


The introduction of a transaction tax system could generally slow down the activities of financial •
investors in commodity markets.


All these measures deserve serious political consideration, even if some of the more sophisticated schemes
among them may prove difficult to implement quickly.






1Motivation of this Study


1. motivation oF this study


Recent developments in commodity prices have
been exceptional in many ways. The price boom be-
tween 2002 and mid-2008 was the most pronounced
in several decades – in magnitude, duration and
breadth. The price decline following the eruption of
the current global crisis in mid-2008 stands out both
for its sharpness and for the number of commodities
affected. Since mid-2009, and especially since the
summer of 2010, global commodity prices have been
rising again. While the recent oil price increases have
been modest compared to the spike in 2007–2008,
food prices reached an all-time high in February
2011. Commodity prices have also been extremely
volatile, in many instances with no obvious link to
changes on the supply side.


Commodity price volatility tends to have signifi-
cant adverse effects. At the macroeconomic level, it
can lead to a deterioration in the balance of payments
and in public finances, and the associated uncertainty
is likely to curtail investment and to significantly
depress long-term growth. At the microeconomic
level, high and volatile commodity prices have severe
impacts on the most vulnerable, especially food- and
energy-insecure households.


Price volatility has long been recognized as a
major feature of commodity markets. Commodity-
specific shocks, especially on the supply side of
food commodities, have generally played a key role
in this respect. Rapidly growing demand for com-
modities, especially in emerging economies, as well
as the debate about the future use of fossil fuels in
the light of global climate change, and about the link


between agricultural production and climate change
more generally, have clearly had an impact on re-
cent commodity price developments beyond simple
commodity-specific shocks. However, since com-
modity prices have moved largely in tandem across
all major categories over the past decade or so, the
question arises as to whether the very functioning of
commodity markets has changed.


A major new element in commodity markets
over the past few years is the greater presence of fi-
nancial investors, who treat commodities as an asset
class. The fact that these market participants do not
trade on the basis of fundamental supply and demand
relationships, and that they may hold, on average, very
large positions in commodity markets, implies that
they can exert considerable influence on the function-
ing of those markets. Indeed, the greater participation
of financial investors may have caused commodity
markets to follow the logic of financial markets more
closely, than that of a purely goods market.


Goods markets may be characterized by an
atomistic market structure and by price discovery
based on information from a multitude of independent
agents who act according to their own individual pref-
erences. By contrast, in financial markets, especially
those whose assets largely fall in the same risk cat-
egory (such as equities, emerging-market currencies
and, recently, commodities), price discovery is based
on information related to a few commonly observable
events, or even on mathematical models that mainly
use past – rather than just current – information for
price forecasts. These differences between goods


Speculators may do no harm as bubbles on a steady stream of enterprise. But the position is
serious when enterprise becomes the bubble on a whirlpool of speculation.


John Maynard Keynes




2 Price Formation in Financialized Commodity Markets: The Role of Information


and financial markets, regarding both the sources of
information and the way information is processed,
imply behavioural differences. In goods markets, the
most profitable market participants will have used
individual, pioneering action based on their own
private circumstantial information. In financial mar-
kets, on the other hand, the most profitable attitude
frequently means following the trend for some time
and disinvesting just before the rest of the crowd does
do so. In other words, a successful financial market
strategy is characterized by herd behaviour. A high
correlation between returns on investment in com-
modities and that in other asset classes is indicative
of such behaviour.


The aim of this study is to provide compre-
hensive insights into recent developments in the
functioning of commodity markets. It pays particular
attention to information flows that affect trading de-
cisions. The study focuses on six commodities: one
energy commodity – crude oil – and five food com-
modities – barley, cocoa, maize, sugar and wheat.


Section 2 of the study focuses on price forma-
tion in commodity markets and also explains the role
of exchanges and over-the-counter (OTC) markets.
Section 3 briefly summarizes recent price develop-
ments and trends of those factors that are commonly
assumed to drive commodity prices – the so-called
“fundamentals”. In this respect, the study focuses
on changes on the demand side, including through


government intervention such as the mandated greater
use of biofuels in some countries. Section 4 addresses
the main focus of the study, namely the increasing
importance of financial investors on commodity
markets. It discusses the institutions, protagonists
and instruments that characterize commodity trading,
as well as the available data on recent commodity
market developments.


Section 5, the other key part of the study,
presents the results of the interviews with physical
and financial traders, as well as other entities involved
in commodity markets. It provides an assessment of
the functioning of commodity markets by market
participants that are involved in commodity trading
on a day-to-day basis.


Based on the analysis in the preceding sections,
section 6 presents policy recommendations. It first
outlines how transparency on physical commodity
markets, as well as on the related futures exchanges
and over-the-counter markets, could be improved.
It suggests that in order to improve the functioning
of commodity futures exchanges in the interests of
producers and consumers, and to keep pace with the
participation of new trader categories such as index
funds, closer and stronger supervision and regulation
of these markets is indispensable. Finally, it ad-
dresses the pros and cons of recently proposed price
stabilization mechanisms. Section 7 concludes with
a summation of the main findings.




3Price Formation in Commodity Markets


This section addresses the main aspects of com-
modity price formation in spot and futures markets.2
It explains the relationship between spot and futures
prices and analyses the role of information in com-
modity markets.


Market participants who need a certain com-
modity at the future time t, can either buy it in the
spot market today and store it, or buy (i.e. take a long
position in) a futures contract and take delivery when
the contract expires. In the former case, the partici-
pants will incur storage costs and opportunity costs
because they might alternatively have invested the
funds used to buy the commodity at the prevailing
interest rate.


The futures price should thus be equal to the spot
price plus interest and storage cost – the so-called
cost of carry. It is expressed as:


F0 = S0 + I + W (1)


F0=futures price at t=0, S0=spot price at t=0,
I=interest, W=storage cost.3


Thus, the price formation of commodity futures
is already linked to financial markets via the inter-
est rate.


If the futures price exceeds the sum of the spot
price and the cost of carry, there is an incentive to
buy the commodity in the spot market and take a
short position (i.e. an obligation to sell the asset) in
a futures contract. This will drive up the spot price
and lower the futures price. As arbitrageurs will be
able to make a risk-free profit as long as F0>S0+I+W,


they buy the commodity in the spot market and sell
a futures contract, engaging in this kind of operation
until prices have adjusted and the futures price is
equal to the spot price plus the cost of carry.


In the opposite case of a lower futures price,
arbitrageurs can sell the commodity on the spot
market, invest the proceeds at the prevailing interest
rate and take a long position (an obligation to buy the
asset) in a futures contract. As long as the arbitrage
possibility persists, a risk-free profit can be made.
Thus arbitrageurs engage in market operations until
the prices have adjusted and equation (1) holds.


In markets for storable commodities, demand
can be met out of current production or inventories.
To the extent that inventories offer protection against
sudden supply disruptions the holder of an inventory
obtains a certain utility from the stock. This utility
is the so-called convenience yield. If inventories are
high, the additional utility from their further increase,
the marginal convenience yield, is low. By the same
token, when inventories are low, and the risk of a
stock running out is high, the marginal convenience
yield from an extra unit of inventories is fairly high.
Thus the marginal convenience yield is inversely
related to inventory levels.


Due to the convenience yield, the forward price
may be below the price defined in equation (1). The
relationship between the futures price and the spot price
– taking the convenience yield into account – is thus,


F0 = S0 + I + W-C (2)


where C is the convenience yield.


2. Price Formation in commodity markets


2.1. information and commodity price formation




4 Price Formation in Financialized Commodity Markets: The Role of Information


In the case of an upward sloping futures curve
(i.e. if futures prices increase with the length of the
maturity of the underlying contract), the market is in
contango. This is typical of situations when inven-
tories are abundant, causing the sum of the storage
cost and the interest rate to exceed the convenience
yield. It implies that the futures price will exceed the
cash price, which is probable, as storage capacity is
limited and therefore storage costs tend to rise with
the level of inventories. This provides an incentive to
sell the commodity on the spot market, which tends
to drive spot prices down.


The opposite, a situation when futures prices
are progressively lower with rising maturity, is called
backwardation. In this case, the futures price does
not cover the cost of carry. Obviously, the demand
for inventories is high in such a situation, because the
convenience yield exceeds the cost of carry. When
inventories are low and the convenience yield is high,
the market is likely to be backwardated, because the
high convenience yield may offset the sum of the
interest and storage costs.


This definition of backwardation should not be
confused with the concept of “normal backwardation”
introduced by Keynes (1930: 142–144). Keynes’ con-
cept refers to an insurance premium paid by hedgers
who take a short futures position (e.g. commodity
producers) and are more risk-averse than their coun-
terparts. Due to this insurance premium, the futures
price exceeds the expected future spot price. Even
in a situation when there are ample liquid stocks of
the commodity and the futures market is in contango,
the expected future spot price still exceeds the futures
price by the premium (i.e. the “normal backwarda-
tion”). However, “normal backwardation” is only
plausible if short hedgers are more risk-averse than
long hedgers, or if the former outnumber the latter.
Numerous studies have tested commodity markets for
“normal backwardation”, with mixed results (see, for
example, Kolb, 1992; and Chang, 1985).


Most commodity markets are characterized by a
low short-run price elasticity of supply and demand.
Consumers have limited substitution possibilities and
substantial medium-term investments are needed, for
example, to develop new oilfields or increase crop
yields. In this market environment, even a compara-
tively small increase in demand leads to substantial
price hikes. The same is true of short-term supply
disruptions, such as those caused by armed conflict
in oil-exporting countries or export bans by grain-


producing countries after a drought. Minor shocks
to quantities will result in significant price reactions.
The situation is exacerbated if inventories are low and
additional demand thus cannot be met out of invento-
ries or, in the case of a supply shock, if releases from
stocks cannot mitigate the price effect.


Due to the high liquidity and easily accessible
information on futures prices, futures markets play
a decisive role in commodity price discovery. The
functioning of this process rests on the EMH. It is
widely believed that the EMH holds in its semi-strong
form, which postulates that any publicly available
information about an asset is reflected in its current
price (see, for example, Malkiel, 1991). Although the
hypothesis initially referred to equity markets, it can
just as easily be applied to prices in other financial
markets as well, such as commodity futures. This
means that any new information on fundamentals of
supply and demand of a commodity leads to a change
in expectations and is immediately incorporated into
commodity futures prices. In the strong form of the
EMH, even private information is reflected in prices.
The reasoning is that as long as information offers
market participants the possibility of a risk-free profit
(arbitrage possibility), they will exploit this oppor-
tunity, causing a movement in the price that reflects
the private information.


The availability of up-to-date and reliable infor-
mation on commodity supply, demand and stocks is
essential for the formation of accurate price expec-
tations and an efficient functioning of commodity
markets. Existing gaps regarding accurate information
on market fundamentals risks causing market partici-
pants to trade on little or wrong information, which
in turn will tend to accentuate price movements and
may cause a sizeable divergence of actual prices
from fundamental values, at least for some period of
time. While information on market fundamentals is
available from a range of sources (see box 1), there
are doubts as to the timeliness and reliability of that
information. Harmonization of data provision and
a more systematic way of data presentation would
greatly facilitate the accessibility of available infor-
mation. Finally, stocks are often held by the private
sector and the proprietary character of the informa-
tion on those stocks causes publicly available stock
data to be particularly incomplete. Owing to these
factors, monitoring and analysing of information on
commodity market fundamentals is a difficult task.
Consequently, a significant proportion of trading in
commodities is subject to considerable uncertainty.




5Price Formation in Commodity Markets


Box 1


sources oF inFormation on commodity market Fundamentalsa


Different types of commodity market information are available, including: (i) raw data from databases that
cover prices, production, consumption, stocks and trade; (ii) processed data based on analyses of market
trends and monitoring of the current situation; and (iii) forecasts or projections of the short- medium- and
long-term evolution of market fundamentals. The frequency of commodity market information varies
widely, depending on the data source, and can range from daily to annual. However, most publicly available
information from official sources is based on monthly data.


There is ample information on physical commodity markets, but it is not easy to obtain in a systematic way.
A number of sources provide the same information, but in different formats. It therefore takes time and
expertise to find out which are the most useful, relevant and reliable sources of information required for a
specific commodity. Even from a single source the multiplicity of information products can make it rather
cumbersome to access the targeted information. The various sources of information include official sources,
such as international organizations and study groups, organizations specializing in specific commodities
or groups of commodities, and governments of countries which are key players in the commodity markets,
such as Australia and the United States, as well as private sources. In many cases, even from official sources,
the information is not publicly available and can be accessed only against payment.


For agricultural commodities, the Food and Agriculture Organization of the United Nations (FAO) is the main
international source for data, market analysis and monitoring of market fundamentals. The FAO publishes
data at different frequencies for various agricultural commodities, most of which can be accessed on the
Internet from its World Food Situation portal. However, a national source, the United States Department
of Agriculture (USDA), is among the most comprehensive sources of information on global agricultural
markets. Its information is particularly important because the United States is a major producing country for a
number of agricultural commodities such as cotton, maize, soybeans and wheat. Therefore, information about
changes in estimations on crops in that country can have a strong impact on global markets. The “Comité du
Commerce des céréales, aliments du bétail, oléagineux, huile d’olive, huiles et graisses et agrofournitures”
(COCERAL) publishes forecasts for grain and oilseed crops for the countries of the EU.


Regarding crude oil, the most comprehensive source of data on production, demand, refinery intake and
output, imports, exports, closing stock levels and stock changes is the Joint Organisations Data Initiative
(JODI). This initiative comprises seven partner organizations: Asia-Pacific Economic Cooperation (APEC),
EUROSTAT, the International Energy Agency (IEA), the International Energy Forum (IEF), the Latin
American Energy Organization (OLADE), the Organization of the Petroleum Exporting Countries (OPEC)
and the United Nations Statistics Division (UNSD). More than 90 countries, representing about 90 per cent
of global oil supply and demand, participate in JODI. The JODI World Database is freely available and
is updated monthly. Information on the major energy consuming countries, is available through the Oil
Market Report online service of the IEA, which provides a monthly assessment of supply, demand, stocks,
prices and refinery activity. On the supply side, OPEC’s Monthly Oil Market Report covers major issues
affecting the world oil market, the outlook for crude oil market developments for the coming year, and a
detailed analysis of key developments impacting oil market trends in world oil demand, supply and the
oil market balance. At the national level, the United States Energy Information Administration provides a
variety of data and analyses on the situation in United States and global energy markets, at different time
frequencies. In the private sector, the widely used annual Statistical Review of World Energy produced by
British Petroleum, provides data about world energy, markets and trends, which are also publicly available.
In addition, Cambridge Energy Research Associates (IHS CERA) is a leading adviser to different clients,
including international energy companies, governments, financial institutions and technology providers. It
delivers critical knowledge and independent analyses on energy markets, geopolitics, industry trends and
strategy.


Platts is a leading global provider of energy information, and among the foremost sources of benchmark price
assessments in the physical energy markets. Argus publishes a full range of business intelligence reports,
market assessments and special studies on all aspects of energy, transport and emissions markets. Commodity
forecasts are also offered by companies specializing in market intelligence, such as the Economist Intelligence
Unit, Business Monitor International and LMC International (agricultural commodities). In addition, the
Working Group on Commodity Prices of the Association of European Business Cycle Institutes (AIECE)
publishes a World Commodity Prices report twice a year, with price forecasts for two years.
This brief review shows that there is an abundance of data sources regarding the fundamentals of physical
commodity markets. Nevertheless, a number of information gaps exist, and there are many areas in which
the transparency of physical commodity markets could be improved, as mentioned in the main text.


a This box is based on Fajarnes (2011).




6 Price Formation in Financialized Commodity Markets: The Role of Information


If prices are driven both by information on fun-
damentals and by factors unrelated to physical supply
and demand in the respective market, the EMH
fails. Price changes may also be due to a “weight-
of-money” effect. This happens when, for example,
index investors take positions that are large compared
to the overall market size. They then face short-term
liquidity constraints, as positions of counterparties
with an interest in the physical commodity are less
than perfectly price elastic. This results in a strong
price impact. However, the price change is not neces-
sarily in line with the fundamentals of the respective
market. Such price movements “in the wrong direc-
tion” may be exacerbated when algorithmic traders
follow the new trend and reinforce it. This is also
likely to occur because many algorithmic traders


use similar models, thus drawing similar conclusions
from market developments.


For these reasons, changes in market prices are
not easy to interpret. Market participants cannot easily
distinguish between price signals that are based on fun-
damentals and contain new information, and distorted
price signals introduced by market participants that trade
on the basis of purely financial news or signals from
mathematical models. As the data based on fundamen-
tals is limited (especially for inventories) it is difficult
to form price expectations. Therefore market partici-
pants may rely, instead, on futures prices to convey the
right signals. This increases the risk of herd behaviour
and a perpetuation of the misleading price signals.
Ultimately it may result in a speculative bubble.


2.2. the role of futures exchanges and otc markets
in commodity price formation


Commodity derivatives are traded either on or-
ganized exchanges or bilaterally “over-the-counter”
(OTC) usually with a financial institution, depending
on the concrete requirements of a trader. However,
a number of exchanges (such as the Intercontinental
Exchange – ICE) also offer OTC transactions and
clearing services.


As the choice of exchange-traded standardized
contracts is limited, there may not be a futures con-
tract which exactly tracks the price developments
of the underlying asset. Differences may be due, for
example, to the delivery point or the quality. Due to
this so-called “basis risk” the standardized contracts
do not always provide a perfect hedge.


This is also why traders tend to choose tailor-
made, non-standardized OTC contracts to hedge their
risks, usually in the form of swaps. OTC contracts
are particularly widely used in energy commodities
such as crude oil or kerosene. A recent report by the
International Organization of Securities Commissions
(IOSCO, 2010) gauges the share of OTC transactions
in all crude oil derivatives at 39 per cent (18 per cent
cleared and 21 per cent uncleared). This means that
substantial counterparty risk is an issue.


Currently, OTC markets in all parts of the world
are still rather opaque, both with respect to the con-
crete positions taken and the way prices are formed.
Price discovery in OTC markets – particularly for
energy commodities – relies heavily on the services
of price reporting agencies (such as Platts or Argus),
which provide thousands of cash reference prices per
day. These benchmarks are commonly used to deter-
mine the floating price component for the settlement
of swaps, though there may be some doubts about the
reliability of these prices (IOSCO, 2010: 5).


In contrast to the OTC market, futures exchanges
trade standardized products with clear definitions of
the quality and quantity of the respective commodity,
and predefined delivery points. Qualities deviating
from these standards or different delivery points are
partly accepted, but traded at a discount. Futures ex-
changes thus offer high liquidity, price transparency
and reduced counterparty risk. Counterparty risk is
limited by the requirement to deposit an initial margin
and settle the account on a daily basis. If the balance
on the account falls below a predefined threshold (the
so-called maintenance margin), a margin call is trig-
gered. The respective market participants then have
to provide additional funds or close their position.




7Price Formation in Commodity Markets


The initial margin is generally only a fraction of the
value of the contract, which means that a trader can
take a position which is several times the value of that
initial margin. Due to the high degree of standardiza-
tion of contracts, exchanges attract a large volume of
trade (i.e. there is high liquidity).


Price developments at the exchanges are im-
mediately reported to news agencies, such as Reuters
or Bloomberg, via the exchanges’ price reporting
systems. There is a high degree of price transparency,
but the positions of various types of traders are only
reported in the United States in an aggregated way
and on a weekly basis. On the whole, prices on futures
exchanges are much more transparent than those in
spot markets, which are comparatively opaque. This
is also emphasized in an IOSCO report (2010: 6)
which states: “The transparency and functioning of
cash markets for commodities remains a prominent
concern.” It is therefore not surprising that futures
markets play such a vital role in commodity price
discovery.


A recent study by the International Food Policy
Research Institute analyses the dynamic relationship
between spot and futures prices of selected agricul-
tural commodities (Hernandez and Torero, 2010).
Using data on weekly returns and weekly volatility
for maize, hard wheat, soft wheat and soybeans, the
study shows that for these commodities, changes
in futures prices lead changes in spot prices more
often than the reverse. The study thus supports the
findings of several earlier ones that reached similar
conclusions. According to the findings of Hernandez
and Torero (2010: 9), “the information flow from
futures to spot markets has intensified in the past
15 years, probably due to the increase in the relative
importance of electronic trading of futures contracts
over open auction trading, which results in more
transparent and widely accessible prices.”


In liquid standardized markets, such as com-
modity exchanges, any substantial price differentials
would not normally be expected to persist for an
extended period, as arbitrage is expected to eliminate
such differentials quickly. The extent price differen-
tials of similar qualities of commodities can persist
also depends on concrete contract specifications. The
most important specification is whether the settlement


is financial or physical, and in the latter case, the rel-
evant delivery points. In the case of physical delivery,
transaction costs, such as transport costs between
delivery points need to be taken into account.


A recent example of a persisting differential
is the wide gap between Brent crude oil prices and
West Texas Intermediate (WTI) oil prices, which
exceeded $15 per barrel in early February 2011. The
price of New York Mercantile Exchange (NYMEX)
WTI, the leading oil contract in the world, has been
significantly below Brent crude futures at ICE, al-
though the two are similar in quality. This can be
explained by various factors. The NYMEX WTI
contract envisages physical delivery in Cushing,
Oklahoma. Inventories in Cushing are soaring, re-
cently reaching a peak of 38.3 million barrels (Meyer,
2011) as a result of increased oil production both in
North Dakota and Canada. As pipelines deliver oil
to Cushing from the north and the south, but can-
not transport oil from Cushing (see, for example,
IntercontinentalExchange, undated), inventories
there keep on rising, whereas demand from nearby
refineries does not keep up, which depresses the
price. Any market participant wishing to engage in
arbitrage would have to move the oil from Cushing
to the Gulf of Mexico for shipment to Europe. This is
costly and takes time. There has been some arbitrage
between Cushing and the Gulf of Mexico to exploit
higher prices on the coast, but the price differential
between WTI and Brent has persisted.


Table 1 offers an overview of relevant exchang-
es for the commodities analysed in this study. It is
difficult to obtain a reliable quantitative ranking of
exchanges by volume, as rankings are usually based
on the number of contracts traded (e.g. by the Futures
Industry Association, FIA). This may be misleading,
because futures contracts for the same commodity
at different exchanges may differ substantially in
size. For instance, the white sugar contract at the
London International Financial Futures Exchange
(LIFFE) refers to 50 tons, whereas the respective
contract at the Zhengzhou commodity exchange
refers to 10 tons. The FIA rankings can provide a
very rough idea as to the relative importance of dif-
ferent exchanges and contracts in the global trade of
commodity derivatives, but not enough to allow any
reliable quantification.




8 Price Formation in Financialized Commodity Markets: The Role of Information


Table 1


leading exchanges For oil and agricultural commodity derivativesa


Exchange Relevant derivatives Relative importance


Chicago Board of Trade
(CBOT) - part of CME Group


Maize, soft red winter wheat - futures, options
wheat-maize inter-commodity spread options


Leading exchange for soft red
winter wheat and maize


Dalian Commodity Exchange
(DCE, China)


Maize - futures Most important exchange for
maize in Asia


Intercontinental Exchange
(ICE)


United States: cocoa, raw sugar (no. 11) -
futures and options
Europe: Brent, WTI - futures and options
Canada: barley - futures and options
OTC: crude oil (various) - swaps


Leading exchange for raw sugar
and cocoa futures (ICE Futures
United States) and Brent crude
oil futures (ICE Futures Europe)


Kansas City Board of Trade
(KCBT)


Hard red winter wheat - futures and options Specialized exchange for wheat


Minneapolis Grain Exchange
(MGEX)


Hard Red Spring Wheat Index (HRSI), Hard
Red Winter Wheat Index (HRWI), Soft Red
Winter Wheat Index (SRWI), National Corn
Index (NCI) - futures and options


Leading exchange for hard red
spring wheat


Multi Commodity Exchange
of India (MCX)


Brent crude oil, crude oil, barley, wheat, feed
maize, white sugar


Among leading exchanges for
crude oil


New York Mercantile
Exchange (NYMEX) - part of
CME Group


Cocoa, raw sugar (No.11) - futures (settlement:
financial)
WTI, Brent, others - futures and options


Leading exchange for light,
sweet crude oil futures;
Among leading exchanges for
other commodities


NYSE LIFFE London: white sugar, cocoa, feed wheat -
futures and options
Paris: milling wheat, malting barley, maize -
futures and options


European exchange for
agricultural commodities


Zhengzhou Commodity
Exchange (ZCE, China)


Hard white wheat, strong gluten wheat, white
sugar - futures


Largest number of contracts for
white sugar, but contract size
is 20 per cent of that at NYSE
LIFFE


Source: Websites of the respective exchanges and Futures Industry Association.
a Concerning the six commodities analysed in this study: barley, cocoa, crude oil, maize, sugar and wheat.




9Recent Evolution of Prices and Fundamentals


In recent years, crude oil prices have climbed
to unprecedented levels, reaching an all-time high of
nearly $150 per barrel in July 2008. In the wake of
the financial crisis of 2008–2009, oil prices fell below
$40 per barrel at the end of 2008 (figure 1).


It is often argued that the fast-growing Asian
emerging economies are a major source of rising
demand for crude oil. The higher energy intensity
of their production compared to that of developed
economies has contributed decisively to the growing
demand (e.g. ECB, 2010). This demand slowed down
only temporarily as a result of the recent crisis.


As Kaufmann (2011) argues, the strong surge in
oil prices in recent years cannot be explained without
taking into account the role of the supply side. There
are two groups of producers in the oil market that
differ significantly in their behaviour. Whereas the
non-OPEC countries can be assumed to be price tak-
ers, with their production positively related to price
and negatively related to cost, the OPEC countries
form a cartel whose operations are based on strategic
considerations. A shift in the supply relations between
the two groups can thus be assumed to have a signifi-
cant impact on the evolution of oil prices. The sudden
slowdown in the growth rate of non-OPEC crude oil
supply after 2004 is therefore seen as a major factor
driving oil price developments (Kaufmann, 2011;
ECB, 2010). It caused an unexpected increase in
OPEC’s capacity utilization, lowering OPEC’s excess
capacity and thus strengthening the role of the cartel
as a marginal supplier.


Recent oil price increases are likely to have been
accelerated by political tensions and armed conflicts
in oil-producing countries, among other factors, al-
though the effect may have been dampened to some
extent by declining inventories. According to the
IEA (2011), current inventories and spare capacity
are still sufficiently high to constrain price increases
in the near future.


3. recent evolution oF Prices and Fundamentals


3.1. crude oil


Figure 1


evolution oF crude oil Prices, 1980–2010
($ per barrel)


Source: UNCTADstat.
Note: The prices shown refer to an equally weighted average


of Brent, Dubai and WTI crude oils.


0


20


40


60


80


100


120


140


19801983198619891992199519982001200420072010




10 Price Formation in Financialized Commodity Markets: The Role of Information


Grain prices have been very volatile4 in the
most recent years. Having peaked in 2008, they
declined sharply, but started rising again in 2010. In
February 2011 maize prices exceeded the level of
June 2008. Due to substitution effects, price move-
ments of the three crops analysed in this study are
highly correlated (figure 2). A number of supply and
demand factors contribute to rising food commodity
prices. Supply growth is slowing, because agricul-
tural land is limited and productivity growth has
slowed (OECD-FAO, 2009). Supply constraints are
exacerbated by the effects of climate change (such
as extreme weather events), which are already felt in
many regions of the world, but are expected to grow
dramatically over the next decades.


On the demand side, the rising world population
and changes in emerging economies towards more
protein-rich diets are major long-term factors. As in-
comes in emerging economies have risen sharply with
accelerated economic growth, consumption patterns


of the population have also changed. Between 1995
and 2005, world meat consumption rose by 15 per
cent, East and Southeast Asia being the region with
the highest increase at almost 50 per cent (FAO,
2009). Taking into account that the production of 1 kg
of meat requires about 7 kg of grains, the impact on
grain demand is substantial.


Biofuel production is another decisive demand
factor. The decision by some governments to intro-
duce blending requirements and subsidies for biofuel
production is considered to play a significant role
in the recent price hikes of grains (box 2). Biofuel
production also affects price movements of agricul-
tural products which are not used in the production
of biofuels, because agricultural land is diverted to
producing crops needed for biofuel production. As
biofuels partly replace petroleum products, they
strengthen the link between the oil market and mar-
kets of agricultural products used in the production
of biofuels (i.e. maize, sugar, oilseeds and palm oil).
High oil prices also affect agricultural commodity
prices via higher production costs, especially for
energy and fertilizers. This may also explain the
co-movement of oil prices and some agricultural
commodity prices.


In the short run, weather effects have a strong
impact on price developments. Often, these are ex-
acerbated by policy measures such as export bans or
taxes. Thus, wheat prices were driven up last August
by the drought in the Russian Federation and an
export ban.


In contrast to grains markets, high and rising
prices are not a new phenomenon in the sugar and co-
coa markets, judged by historical standards (figure 3).
These two soft commodities already experienced ex-
treme price spikes in the 1970s and 1980s. Recently,
the cocoa price has come under pressure due to po-
litical tensions in Côte d’Ivoire, the world’s largest
cocoa producer. The sugar price has risen sharply
despite production increases. Expected higher de-
mand may be a factor (FAO, 2010).


3.2. selected food commodities


Figure 2


evolution oF grain Prices, 1980–2010
($ per ton)


Source: UNCTADstat; and IMF, primary commodity price tables.


0


50


100


150


200


250


300


350


400


450


500


1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010


Barley


Maize


Wheat




11Recent Evolution of Prices and Fundamentals


Box 2


BioFuelsa and their role in driving uP commodity Prices


In recent years, a number of countries have introduced or expanded mandates for the blending of fossil
fuels with biofuels.b These requirements have been driven largely by political rather than economic motives,
based on the need to reduce greenhouse gas emissions, increase energy independence and support rural
incomes.c In addition to blending requirements, biofuel production is supported by high subsidies, the highest
being in the United States, where almost $6 billion were spent in support of biofuels in 2006 (Steenblik,
2007). Biofuels are heavily subsidized also in the EU. Brazil has the highest ethanol blending requirement,
at between 20 and 25 per cent, but its ethanol is produced from sugar cane which is competitive without
subsidies (FAO, 2008).


Spurred by subsidies and blending requirements in a
number of countries, as well as rising crude oil prices,
biofuel production has increased more than fivefold since
2000 (see figure). However, biofuels still account for
only about 2 per cent of world oil supply and therefore do
not yet affect crude oil prices. Biodiesel, the production
of which started from a much lower level, increased its
share in world biofuel production from 5 per cent in
2000 to 18.8 per cent, but the share of ethanol continues
to be higher. There are substantial regional differences.
Production of biofuels is heavily concentrated in the
United States and Brazil, which accounted for 45.6 per
cent and 29.2 per cent, respectively, of total biofuel
production in 2009, while the combined share of
Europe and Asia was just 20 per cent (see also annex
table A.1).


The world’s largest ethanol producer, the United States,
almost exclusively uses maize for ethanol production.
According to data from the USDA, the share of the total


maize production which is used for ethanol production has almost doubled since 2006 and is now close
to 40 per cent. The period of the most rapid expansion of United States ethanol production coincides with
strong increases in grain prices. In contrast, Brazil bases its ethanol production on sugar cane, of which it is
the world’s leading producer, using 55–60 per cent of its sugar cane production to produce fuel (McConnell,
Dohlmann and Haley, 2010). Owing to the strong increase in ethanol production, Brazil’s output of sugar
cane has risen fast, albeit more slowly than its ethanol production. There are substantial differences in the
so-called fossil energy balance of biofuels (i.e. the ratio of energy contained in biofuel to fossil energy used
in its production). Whereas the fossil energy balance of ethanol produced from maize is less than two, that
of ethanol produced from sugar cane ranges between 2 and 8 (FAO, 2008: 17).


A recent study by UNCTAD (2009a: 1) estimates that, due to blending requirements in many countries,
demand for biofuels will rise much faster than production capacity. In addition, subsidization of biofuels
implies that biofuel production has zero elasticity with respect to changes in feed prices. For these reasons it
seems plausible that enhanced biofuel production has had some effect on maize prices and – via substitution
effects – also on prices of other grains such as barley, rice and wheat. In addition to the direct price effects
of higher demand for those crops which serve as feedstock for biofuel production, there are also indirect
price effects on other crops which result from changes in land use in favour of crops for biofuel.


A number of studies find significant effects of biofuel production on agricultural commodity prices. For
example, growing biodiesel production in Europe has indirectly exacerbated price rises in the wheat
market, because land which would otherwise have been used for growing wheat has been diverted to
oilseed production (Mitchell, 2008). It has also had an effect on other food products (such as meat and
dairy products), which require the same agricultural commodities as a production input (Helbling, Mercer-
Blackman and Cheng, 2008). Estimates of the effects of biofuels on maize prices range from 39 per cent


World BioFuel Production
(Thousand barrels per day)


Source: Energy Information Agency (EIA), International
Energy Statistics database.


/...


0


200


400


600


800


1 000


1 200


1 400


2000 2003 2006 2009


Biodiesel


Ethanol




12 Price Formation in Financialized Commodity Markets: The Role of Information


Unlike in earlier periods, the recent price hikes
have occurred in an environment of general price
increases across a wide range of commodities, from
energy to agricultural commodities. Most of the
factors which are often cited as price drivers, such
as population growth or changing consumption pat-
terns have been at work for an extended period often
coinciding with low commodity prices. Their role in
explaining recent price hikes is therefore doubtful.
Experiences with the weak forecasting performance
of econometric models for oil prices based on fun-
damentals (e.g. Kaufmann, 2011) also suggest that
physical supply and demand are not the only factors
that drive oil prices. The European Commission
(2008) has also expressed doubts that market funda-
mentals are the main drivers of commodity prices. As
the following sections show, there is strong evidence
that the increasing presence of financial investors in
commodity markets plays an important role in price
dynamics.


Figure 3


evolution oF Prices oF selected soFt
commodities, 1980–2010


(US cents per pound)


Source: UNCTADstat.


(Rosegrant, 2008 for the period 2000–2007) to between 70 and 75 per cent (Mitchell, 2008 for the period
2002–2008). Roberts and Schlenker (2010) estimate the impact of United States biofuel production alone
on world prices of maize, rice, soybeans and wheat to be about 30 per cent. If a third of the calories used
were recycled to feed animals, the price effect would still be 20 per cent. For rice and wheat, the price
effects may be slightly lower than for maize, according to Rosegrant (2008) who estimates the impact at
21 per cent and 22 per cent, respectively, of price increases between 2000 and 2007.


Several other studies find no significant effects, or argue that biofuels cannot have serious effects on
agricultural commodity prices. Baffes and Haniotis (2010) contend that it is highly unlikely that biofuel
production triggered recent agricultural commodity price spikes, given the small share of land used for
biofuels in global land used for grain and oilseed production. Based on an analysis of data on food commodity
prices, production, inventories and trade, Pfuderer and del Castillo (2008) conclude that biofuel production
has not been the main driver of recent commodity price hikes. However, their analysis leaves some open
questions. For example, in their analysis of the wheat price, little account is taken of substitution effects
between various grains. Trostle (2008) acknowledges that United States ethanol production, which accounts
for 30 per cent of the global growth in wheat and feed grain consumption, had some effect on world markets.
On the other hand, he stresses that the effect was mitigated by the availability of by-products of ethanol
production (so-called “distillers’ grains”) for feed purposes and by an increase in land use.


On balance, the evidence supports the view that biofuels have contributed to the recent increase in food
prices but estimates as to the extent of this effect diverge widely.


a As this study focuses on grains (barley, maize, wheat), cocoa, crude oil and sugar, the use of oilseeds for biodiesel
is not highlighted here.


b Pfuderer and del Castillo (2008) provide an overview (see also FAO, 2008: 29).
c More recently, however, doubts have emerged about the environmental benefits of biofuels. Taking the change of


land use into account the net benefit of biofuels for the reduction of greenhouse gases might actually be negative
(see e.g. Searchinger, 2008).


Box 2 (concluded)


0


20


40


60


80


100


120


140


160


180


1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010
0


5


10


15


20


25


30


35


40


45


Cocoa (left scale)


Sugar (right scale)




13Financialization of Commodity Price Formation


The term “financialization of commodity trad-
ing” indicates the increasing role of financial motives,
financial markets and financial actors in the operation
of commodity markets.


Investors have been engaging in commodities
trading for the purpose of portfolio diversification
ever since it became evident that commodity futures
contracts exhibited the same average returns as in-
vestments in equities, while over the business cycle
their returns were negatively correlated with those on
equities and bonds. The empirical evidence for this
finding emerged from an analysis of data stretching
over a long period, from 1959 to 2004 (Gorton and
Rouwenhorst, 2006). That analysis also shows that
the returns on commodities were less volatile than
those on equities or bonds, because the pair-wise
correlations between returns on futures contracts
for various commodities (e.g. oil and copper, or
oil and maize) were relatively low (Gorton and
Rouwenhorst, 2006).


Commodity futures contracts were also found
to have good hedging properties against inflation (i.e.
their return was positively correlated with inflation).
This is because they represented a bet on commodity
prices, such as prices of energy and food products,
which have a strong weight in the goods baskets
used for measuring current price levels. Also, futures
prices reflect information about expected changes in
commodity prices, so that they rise and fall in line
with deviations from expected inflation.


Furthermore, investing in commodity futures
contracts may provide a hedge against changes in the


exchange rate of the dollar. Most commodities are
traded in dollars and commodity prices in dollar terms
tend to increase as the dollar depreciates. Measured
in a currency basket, commodity prices are generally
less correlated with the dollar; indeed, the sign of the
correlation is reversed (IMF, 2008: 63). This suggests
that changes in the value of the dollar against other
currencies may partly explain the negative correlation
between the prices of dollar-denominated commodi-
ties and the dollar.


Financial investors have long been active on
commodity markets.5 But the above mentioned
empirical findings of their investments in commodi-
ties for purposes of portfolio diversification gained
considerable attention following the bursting of the
equity market bubble in 2000, which spurred financial
investment in commodities.6 Moreover, there was
growing acceptance of the notion that commodities
as an asset class are a quasi-natural hedge against po-
sitions in equity markets (Gorton and Rouwenhorst,
2006), as already mentioned.


Such portfolio diversification considerations
gained further impetus in the early 2000s with the
increasing recognition in both academic circles
(e.g. Radetzki, 2006) and among potential investors
(e.g. Heap, 2005) that commodities were entering a
new super cycle. It was believed that rapidly growing
demand associated with urbanization and industriali-
zation, as well as changes in dietary habits towards
more protein-rich diets in major emerging economies,
particularly China and India, had triggered a new,
prolonged increase in real commodity prices (see
also UNCTAD, 2005).


4. Financialization oF commodity Price Formation


4.1. Financialization: definition, motivation, size and instruments




14 Price Formation in Financialized Commodity Markets: The Role of Information


Financial investors use a range of instruments.7
However, investment in commodity indexes has
probably attracted the most attention over the past
few years. Index investment tracks returns on weight-
ed commodity baskets (e.g. the Standard & Poor’s
Goldman Sachs Commodity Index (S&P GSCI)
and the Dow Jones-Union Bank of Switzerland
Commodity Index (DJ-UBSCI)).8 These indexes
are composites of futures contracts on a broad range
of commodities (including energy products, agri-
cultural products and metals) traded on commodity
exchanges. Investing in a predetermined basket of
commodities, as is done in index investment, rests on
the assumption that commodities have a unique risk
premium which is not replicable by combining other
asset classes, and that they form a fairly homogeneous
class which can be represented by a few positions
(Scherer and He, 2008). These characteristics are
likely to be accentuated in periods of commodity
super cycles. During those periods, commodity-
specific market intelligence, as generally gathered
by investors that focus on specific commodities,
may be considered unnecessary. As a result, the fees
associated with investing in commodity indexes are
fairly low.


Financial investors gain exposure in commodity
indexes by entering into a bilateral financial agree-
ment, usually a swap, with a bank or another large
financial institution. The investor purchases parts
in a commodity index from the bank, and the bank
in turn hedges its exposure resulting from the swap
agreement through futures contracts on a commod-
ity exchange. Financial investment in commodity
indexes involves only “long” positions (i.e. pledges
to buy commodities) and relates to forward posi-
tions (i.e. no physical ownership of commodities is
involved at any time).9 This process – known as “roll-
ing” – gives rise to a roll yield which is positive in a
“backwardated” market and negative in a “contango”
market.10 This specific characteristic of index trading
implies that roll yields are of particular importance
to position-taking by index traders.


Financial investors that follow a more active
trading strategy, such as money managers (see be-
low), are unlikely to rely on long-term oriented index
investment; rather, they tend to operate on the basis of
more short-term investment horizons and take posi-
tions on both sides of the market through futures and
options contracts. This enables them to earn positive
returns in both rising and declining markets.


Since about 2009, a third basic instrument has
gained considerable importance, namely so-called
“exchange-traded products” (ETPs). Most ETPs,
which comprise exchange-traded notes (ETNs) and
exchange-traded funds (ETFs), replicate the return
on a single commodity, while a few track com-
modity groups. The shares of ETPs are traded on
equity markets. Some of them are easily accessible
by small-scale investors, while others offer large
single coupons and are therefore more attractive to
institutional investors such as pension funds. Apart
from ETFs for precious metals, such funds have
traditionally used futures contracts as collateral. But
an important recent development is that some ETPs,
such as those in copper and aluminium, are backed by
physical commodities. Futures-backed ETPs expose
investors to counterparty risk, as transactions involv-
ing buying or selling of ETPs do not go through a
clearing house on commodity exchanges. The rising
importance of physically-backed ETPs indicates that
risk aversion and growing concern with counter-
party risk have made it more acceptable for financial
investors to bear the storage cost of the physical
commodities as they can be used as collateral. The
currently very low interest rates, which reduces the
cost of credit used to finance storage costs, has most
likely also contributed to the increased importance of
physically-backed ETPs. Returns on such products
are determined by spot price movements, while the
returns on futures-backed ETFs are largely influenced
by the roll yield, and thus share the characteristics of
traditional index investments.


The further expansion of physically backed
ETPs may well cause a tightening of physical
commodity supply, because part of the physical
commodities available in the warehouses of com-
modity exchanges will be earmarked as collateral,
and therefore will not be available for delivery. This
could give rise to a cash premium (or increase exist-
ing premiums) and move commodity markets into
backwardation, which in turn would increase the
return on commodity index investments and make
such investments more attractive. To the extent that
this would cause an increase in commodity prices, it
would increase the need for physically backed ETPs
in order to hold more physical commodities. In other
words, the conjunction of these two instruments may
well ignite a speculative bubble. Moreover, due to
the close link of returns to spot price movements, an
increasing popularity of physically backed commod-
ity investments would most likely exacerbate price




15Financialization of Commodity Price Formation


volatility, as investors would buy such instruments
in times of rising prices, but sell in times of declin-
ing prices.


Financial investors are also increasingly using
structured products. These products can take differ-
ent forms, but typically combine an underlying asset
with a derivative (such as an option). The addition of
a derivative is often aimed at protecting the capital
invested in the underlying asset, and thereby reducing
risk while maintaining the possibility of benefiting
from the current price trend. This option gives the
right (but not the obligation) to buy (or sell) an asset at
a specified price within a given time frame. Given the
generally non-standardized character of structured
products, they are typically traded OTC between an
investment bank and a financial investor. Structured
products on commodity indexes first appeared on the
market in 2006. They compete with the traditional,
broad-based commodity indexes but make the rolling
process more flexible, thereby reducing roll losses.


It is difficult to assess the size of the finan-
cialization of commodity trading due to the lack of
comprehensive data. But it is reflected, for example,
by the strong increase, starting around 2004, in the
number of futures and options contracts outstand-
ing on commodity exchanges and in the amount


of outstanding OTC commodity derivatives. The
number of contracts outstanding on commodity ex-
changes has continued to increase since the collapse
of commodity prices in mid-2008, and is now about
50 per cent higher than in the first half of 2008, when
commodity prices peaked (figure 4). In contrast, the
notional amount of outstanding OTC-derivatives
has dropped to about one third, which corresponds
to roughly half of its level in 2005–2006, but also to
about five times its level in 1999 (figure 5).11


A number of reasons could explain the sharp
decline in the notional value of outstanding OTC
commodity derivatives. The collapse of commodity
prices between mid-2008 and early-2009 to about
half their previous level clearly accounts for part of
this decline.12 Another reason could be that the finan-
cial crisis led to greater awareness of counterparty
risk, making financial investors wary of exposure
in bilateral OTC deals. Thirdly, the recent fall in
recorded OTC activity probably reflects a decline
in the relative importance of broad-based passive
index investments by financial investors in com-
modities, which includes the use of swaps on OTC
markets, and an increasing relative importance of
more sophisticated active trading strategies, which
emphasize the use of futures contracts traded on
organized exchanges. A survey conducted in early


Figure 4


Futures and oPtions contracts
outstanding on commodity exchanges,


decemBer 1993–decemBer 2010
(Number of contracts, millions)


Source: Bank for International Settlements (BIS), Quarterly
Review, March 2011, table 23B.


Figure 5


notional amount oF outstanding
over-the-counter commodity derivatives,


decemBer 1998–June 2010
(Trillions of dollars)


Source: BIS, Quarterly Review, March 2011, table 22A.


0


10


20


30


40


50


60


70


1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
0


2


4


6


8


10


12


14


1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2008 2009 2009 2010


Other commodities
Other precious metals
Gold


(June) (June) (June)




16 Price Formation in Financialized Commodity Markets: The Role of Information


December 2010 on how commodity investors plan
to invest in the coming 12 months indicated that only
7 per cent expected to use index swaps compared with
43 per cent that would choose active management
(Barclays Capital, 2010). Such active management
includes the use of ETPs, such as ETFs, which may
be backed by futures contracts.


Evidence on the value of assets under manage-
ment by financial investors in commodities reveals
two salient features (figure 6). First, these investors
have rapidly increased their involvement in com-
modities even more since mid-2010 than before the
financial crisis when it was already growing fast.
Judging from currently available data, the commod-
ity-related assets under their management recorded
a historic high in March 2011, when it reached about
$410 billion – about double the pre-crisis level of
2007. Second, while index investment accounted for
65–85 per cent of the total between 2005 and 2007
prior to the financial crisis, its relative importance
has fallen to only about 45 per cent since 2008.
This decline occurred despite a roughly 50 per cent
increase in the value of index investments between
2009 and the end of 2010.


To put the size of financial investments in
commodities in perspective, it is useful to consider
how these have evolved relative to investments in


equity markets, and relative to developments in the
real economy. Between about 2002 and the out-
break of the financial crisis, the notional amount of
outstanding OTC commodity derivatives increased
considerably faster than comparable investments in
equity-linked contracts. However, in 2008–2009 the
value of commodity investments also declined con-
siderably faster than that of equity-linked investments
(figure 7). Perhaps more importantly, the share of
the notional amount of outstanding OTC commodity
derivatives in global gross domestic product (GDP)
increased from 2–3 per cent in the early 2000s to more
than 20 per cent in 2008, and, in spite of its subsequent
rapid decline, this share has remained at about 5–6 per
cent (i.e. roughly double its share about a decade
ago). The evidence in figure 7 also reflects the dif-
ferences in the evolution of commodity investments
on exchanges and on OTC markets, noted above; it
shows that the share of the value of commodity assets
under management in global GDP increased more
than fourfold in the period 2008–2010.


A comparison of the development of physical
commodity production and financial investment
in commodities sheds some further light on the
size of the financialization of commodity markets.
Concentrating on oil, which constitutes the largest
share of total commodity production, reveals that the
share of the notional value of total (i.e. not just oil


Figure 6


Financial investments in commodities: assets under management, By Product, 2005–2011
($ billion)


Source: Barclays Capital, The Commodity Investor, various issues.


0


50


100


150


200


250


300


350


400


450


2005 2006 2007 2008 2009 2010
2nd quarter


2010
3rd quarter


2010
4th quarter


2011
1st quarter


Barcap estimates of indices Medium-term notes ETP-indices ETP-precious metals
ETP-industrial metals ETP-energy ETP-agriculture


2010
1st quarter




17Financialization of Commodity Price Formation


for which no separate data are available) outstanding
OTC commodity derivatives in the value of global
oil production increased about fourfold between the
early 2000s and 2007–2008 when it reached 40–45
per cent (shown by the dark columns in figure 8).


A similar value-based measure relating to financial
investments in commodity futures exchanges shows
that the share of the notional value of the outstand-
ing index investments in WTI crude oil on United
States futures exchanges in the value of global oil


Figure 7


Financial investments in commodities and gloBal gdP, 1998–2010
(Per cent)


Source: UNCTAD secretariat calculations, based on BIS; Barclays Capital; and UNCTADstat.


0


5


10


15


20


25


1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
0.0


0.1


0.2


0.3


0.4


0.5


0.6


0.7


0.8


0.9


1.0


Notional value of outstanding OTC equity-linked derivatives as a share of global GDP (left scale)
Notional value of outstanding OTC commodity derivatives as a share of global GDP (left scale)
Commodity assets under management as a share of global GDP (right scale)


Figure 8


Financial investments in commodities and gloBal oil Production, 2001–2010
(Per cent)


Source: UNCTAD secretariat calculations, based on BIS; CFTC; IEA; and UNCTADstat.


0
5


10
15
20
25
30
35
40
45
50


3 6 9 12 3 6 9 12 3 6 9 12 3 6 9 12 3 6 9 12 3 6 9 12 3 6 9 12 3 6 9 12 3 6 9 12 3 6 9


2001 2002 2003 2004 2005 2006 2007 2008 2009 2010


0.0


0.5


1.0


1.5


2.0


2.5


3.0


Notional value of outstanding OTC-commodity derivatives as a share of value of global oil production
(left scale)
Notional value of outstanding index investment in WTI crude oil on United States futures markets as a
share of value of global oil production (left scale)
Number of commodity contracts traded on organized exchanges as a share of barrels of global oil
production (right scale)




18 Price Formation in Financialized Commodity Markets: The Role of Information


production in 2010 was about 50 per cent higher
than in 2007–2008 (shown by the light columns in
figure 8). Given that WTI appears to have ceded
part of its function as a benchmark for global crude
oil prices to Brent, this increase may well be an


underestimation. Indeed, comparing the number of
commodity contracts traded on organized exchanges
and the volume of global oil production (indicated by
the line in figure 8), indicates an unabated increase in
the financialization of commodity markets.


4.2. categories of market participants


Several categories of market participants are
active in commodity markets.13 These categories are
usually distinguished on the basis of the reports on
traders’ positions that are published in anonymous
and summary form by the CFTC – in its weekly
Commitment of Traders (COT) reports. The main
purpose of these reports is to improve transparency
about activity in futures markets.


The CFTC used to distinguish only between
two categories of market participants: those that
hedge an existing exposure, which it categorized as
“commercial”, and those that do not hedge which
it categorized as “non-commercial”.14 However, it
became widely perceived that, as a consequence of
the growing diversity of market participants in fu-
tures exchanges and the greater complexity of their
activities, the traditional COT data may fail to fully
reflect such activity (CFTC, 2006). This is because
those hedging, and therefore defined as commercial
market participants, have normally been considered
as entities that use transactions in futures contracts
to reduce risk in the conduct of a commercial enter-
prise. However, many market participants who report
positions as hedges, and who therefore fall under
the “commercial” category, are in fact commodity
swap dealers, who hedge to offset financial posi-
tions. If their underlying positions were held directly
as commodity futures contracts (rather than being
intermediated through OTC swap agreements), they
would be categorized as “non-commercial”.


Responding to these concerns, in 2007 the
CFTC introduced a new and better categorization in
its Supplementary Commodity Index Traders (CIT)
reports with data on positions of index traders for
12 agricultural commodities.15 The index trader
positions include those taken by both pension
funds, previously classified as non-commercial


traders, and swap dealers that had been classified as
commercial traders. According to the CFTC (2009),
CITs generally replicate a commodity index, but may
belong to either the commercial or non-commercial
category.


In September 2009, the CFTC went even further
and started to publish Disaggregated Commitment of
Traders (DCOT) reports. These reports have been
providing weekly data beginning in June 2006 for
the 12 agricultural commodities covered by the CIT
reports plus a range of energy commodities and met-
als, such as crude oil, natural gas, copper and gold.
The DCOT reports distinguish five trader categories
(see table 2).


The DCOT reports consider the first two trader
categories (i.e. PMPU and swap dealers) as “com-
mercial” traders, and the other two reporting trader
categories as “non-commercial” traders. By contrast,
the index trader category of the CIT reports does not
coincide with the swap dealer category in the DCOT
reports. This is because the swap dealer category of
the DCOT reports includes swap dealers who do not
have commodity index-related positions, and there-
fore are not included in the index trader category of
the CIT reports. Also, the index trader category of the
CIT reports includes pension and other investment
funds that place index investments directly into the
futures markets rather than going through a swap
dealer; these traders are classified as managed money
or other reportables in the DCOT reports (see also
Irwin and Sanders, 2010).


Money managers generally have a short-term
perspective and adopt an active investment strategy.
This strategy goes beyond the consideration of com-
modities as a fairly homogenous asset class with
a unique risk premium, which is characteristic of




19Financialization of Commodity Price Formation


broad-based passive index investment; it also takes
into account factors such as different short-term
supply-and-demand dynamics, as between industrial
metals and energy. Perhaps more importantly, active
trading strategies try to take advantage of profitable
investment opportunities arising: (i) in declining
markets (by taking “short” in addition to “long” posi-
tions); (ii) from taking longer dated futures positions
than those usually included in readily available in-
dexes; (iii) from trading commodities that are barely,
if at all, included in the popular commodity indexes
(e.g. soybean oil is not included in the S&P GSCI,
while cocoa is not included in the DJ-UBSCI); and
(iv) from employing a “relative value” approach,
such as by exploiting differences in quality (e.g. WTI
versus Brent crude oil), regional dynamics (e.g. North


America and Western Europe versus Asia), intra-
commodity dynamics (e.g. soybeans versus soybean
oil), and cross-commodity dynamics (e.g. trading oil
and feedstock used for biofuel production against
other food commodities).


The money manager category includes a range
of investors, such as hedge funds and institutional
investors, which follow different trading strategies
based on macroeconomic fundamentals, detailed
commodity research, algorithmic trading or trend fol-
lowing, and general financial portfolio-diversification
considerations. Thus they are able to adjust their ex-
posure in commodity markets according to changes
in asset prices with a view to stabilizing the structure
of their portfolio.


Table 2


trader categories in the cFtc’s disaggregated commitment oF traders rePorts


1. Producers, merchants,
processors, users (PMPU)


Entities that predominantly engage in the physical commodity markets and
use the futures markets to manage or hedge risks associated with those
activities.


2. Swap dealers Entities that deal primarily in swaps for a commodity and use the futures
markets to manage or hedge the risks associated with those swap
transactions. The bulk of these traders’ clients are index investors who invest
in commodity indexes such as the S&P GSCI and the DJ-UBSCI.


3. Money managers Entities that manage and conduct organized futures trading on behalf of
their clients. This category includes registered commodity trading advisers
(CTAs), registered commodity pool advisers (CPOs), and unregistered funds
identified by the CFTC. Hedge funds and large ETFs are part of this category.


4. Other reporting traders Every other reportable trader that is not included in one of the other three
categories.


5. Non-reporting traders Smaller traders who are not obliged to report their positions.




20 Price Formation in Financialized Commodity Markets: The Role of Information


The financialization of commodity trading
has made the functioning of commodity exchanges
controversial. Their traditional functions have been
to facilitate price discovery and allow the transfer
of price risk from producers and consumers to other
agents that are prepared to assume the price risk. These
functions are impaired to the extent that trading by
financial investors increases price volatility and drives
prices away from levels that would be determined by
physical commodity supply and demand relation-
ships. As a result, commodity price developments
no longer merely reflect changes in fundamentals;


they also become subject to influences from financial
markets. Consequently, market participants with a
commercial interest in physical commodities (i.e.
producers, merchants and consumers) face greater
uncertainty about the reliability of signals emanating
from commodity exchanges. Thus, managing the risk
of market positions and making storage, investment
and trading decisions become more complex. This
may discourage long-term hedging by commercial
users. Moreover, with greater price volatility, hedging
becomes more expensive, and perhaps unaffordable
for developing-country users, as well as riskier.16


4.3. What is problematic about financialization?


The availability and processing of information
plays a key role in the determination of asset prices.
This role has traditionally been examined on the basis
of the EMH, whereby prices perfectly and instantane-
ously respond to all available information relevant to a
freely operating market. Market participants continu-
ously update their expectations from inflowing public
and private information. This means that prices will
move either when new information becomes publicly
available (in the case of commodities, for example
following announcements of harvest forecasts or
changes in oil production), or when private informa-
tion leads to transactions that affect prices.


Crucial assumptions of the EMH are that market
participants evaluate assets on the basis of funda-
mentals, act fully rationally, base their actions on
publicly available or their own private information,
and do so independently of each other. However,
some circumstances can cause individuals to deviate
from this assumed behavioural pattern and to engage
in herd behaviour. Herd behaviour frequently occurs
when decisions need to be taken in situations of
uncertainty.17 It may be defined as the tendency of


individuals to mimic the actions of a larger group,
rather than acting independently and on the basis of
their own information.


Herd behaviour can take various forms and may
be rooted in irrational behaviour, but it may also be
fully rational. Figure 9 provides a taxonomy of dif-
ferent types of herd behaviour. Early models of herd
behaviour were based on assumed deviations from per-
fect rationality, or so-called “noise trading” (Shleifer
and Summers, 1990). Investment by noise traders is
affected by pseudo-signals, which convey no informa-
tion about future returns in a specific asset market, or
by changes in traders’ beliefs and sentiments that are
not justified by news on fundamentals. An example
of pseudo signals for positions in commodity mar-
kets is information related to other asset markets that
triggers portfolio rebalancing, and, hence, changes
in investors’ exposures to commodities.


Changes in beliefs and sentiments may reflect
investors’ judgemental biases, such as overreacting
to news or overoptimism.18 It may also reflect use
of inflexible trading strategies, such as momentum


4.4. Herd behaviour and the limits of arbitrage




21Financialization of Commodity Price Formation


investment or positive feedback strategies. Such
strategies assume that past price developments
carry information on future price movements giving
rise, for example, to trend chasing. This will result
in buying after prices rise and selling after prices
fall, independently of any changes in fundamentals.
Simple types of positive feedback strategies are
closely related to technical analysis that utilizes past
price and position data to assess patterns of activity
that might be helpful in making predictions. More
sophisticated trading rules use computer-based al-
gorithms that strictly adhere to a predetermined set
of rules. Algorithms analyse market activity and pro-
duce signals for trading strategies established either
on the basis of past trading and price developments
or on the basis of the anticipated reaction by other
algorithmic traders to current market developments.19
Given that several positive-feedback and algorithmic
traders may use similar rules, they run the risk of col-
lectively generating market movements that they then
individually identify and follow. Moreover, to the
extent that algorithms follow statistical strategies and
monitor market developments across different asset
markets, such rules will cause price signals to spill
over from, for example, equity or currency markets
to commodity markets, even when there is no change
in the fundamentals on commodity markets.


Herd behaviour can also be fully rational. In
this context, “spurious herding” should be distin-
guished from “intentional herding” (Bikhchandani
and Sharma, 2001). Spurious herding describes situ-
ations where agents facing similar decision-making


problems and information sets take similar decisions.
Given that spurious herding reflects agents’ common
reaction to public information, it is entirely compat-
ible with the EMH, provided that the information
refers to the fundamentals of the specific market.
Fundamentals-driven spurious herding in commod-
ity investment can arise if, for example, a significant
share of international supply is suddenly cut off, as
occurred with oil during the Gulf war in 1990–1991
and with rice following the imposition of export bans
by various large exporting countries in 2008.


Intentional herding may be based on four mo-
tives (Devenow and Welch, 1996; Bikhchandani
and Sharma, 2001). First, conformity-based herding
relates to an alleged intrinsic preference of individu-
als for conformity. Second, reputation-based herding
relates to imitation which arises when traders and
their employers are uncertain about the traders’ abili-
ties (Scharfstein and Stein, 1990). Traders who doubt
their own abilities will not take positions contrary to
those taken first by other traders, even if their own in-
formation would lead them to do otherwise. Doubtful
traders, by imitating others, will avoid being considered
low-skilled if taking positions contrary to those taken
by others turned out to be loss-making. If the common
decision turns out to be loss-making, it will be attrib-
uted to a change in general market sentiment, rather
than to poor individual judgement or performance.20
Third, closely related to reputation-based herding is
compensation-based herding. This refers to agents who
invest on behalf of others and whose compensation
schemes and terms of employment provide incentives


Figure 9


diFFerent tyPes oF herd Behaviour


Source: UNCTAD secretariat, derived from Bikhchandani and Sharma (2001); and Shleifer and Summers (1990).


Information-
based


Reputation-
based


Spurious
herding


Conformity-
based


Beliefs/sentiments/
positive feedback


strategies/algorithms


Rational
herding


Intentional
herding


Compensation-
based


Irrational
herding


Noise
trading


Pseudo
signals




22 Price Formation in Financialized Commodity Markets: The Role of Information


that reward imitation. For example, risk-averse
investors will align their positions with benchmark
portfolios if their compensation increases when they
do better than the benchmark but decreases when
they underperform the benchmark. Compensation
rules based on such relative performance measures
can lead not only to herding but also to risk-loving
investors taking excessively high risk.


Fourth, information-based herding is perhaps
the most important motive of intentional herding. It
refers to imitation in situations where traders believe
that they can glean information by observing the
behaviour of other agents. In other words, investors
converge in their behaviour because they ignore their
private information signals (Hirshleifer and Teoh,
2003). As explained by Banerjee (1992), who calls
this effect “herd externality”, information-based
herding exerts an external influence on decision-
making processes and causes position-taking that is
not in line with an agent’s own information. Position-
taking based only on other peoples’ previous actions
will cause price changes without infusing any new
information into the market. A sequence of such
actions causes a so-called “informational cascade”
(Bikhchandani, Hirshleifer and Welch, 1992) – a
snowballing effect which will eventually lead to self-
sustaining asset price bubbles.


Informational cascades are most likely to occur
where market participants are unequally informed and
ignore the accuracy of other peoples’ information.
Market participants who judge their own informa-
tion to be incomplete and approximate will tend to
delay their decision-making, preferring to act only
once they can make inferences on the basis of other
– supposedly better informed and more experienced
– people’s action. This implies that position-taking
by investors that make early decisions is likely to
determine which way followers will decide to move,
and it therefore has a disproportionate impact on
price changes. This will be the case even if the as-
sessments of the early movers are incorrect, based
on overconfidence or on idiosyncratic motives (such
as readjusting portfolio composition following price
changes in other asset markets). It also implies that
an increase in the number of market participants and
in liquidity does not necessarily indicate that market
transactions are based on more information.


Informational cascades are not limited to one
market. They can spread across different asset mar-
kets if prices in those markets are correlated. Herding


across markets can lead to excess correlation (i.e. a
level of correlation between asset prices that exceeds
the correlation between their fundamentals) (Cipriani
and Guarino, 2008).


Informational cascades and information-based
herding can be altered or even reversed by a publicly
observable shock or by the release of public informa-
tion (Hirshleifer and Teoh, 2003). Both events infuse
new information into the market. They also allow
followers to assess the accuracy of the information on
which they assumed precursors were acting, as they
know that the newly released public information is
more accurate than what they inferred from the ac-
tions of the early position-takers. Such new public
information may consist of easily observable events
(such as extreme weather events that impact harvests)
or well-researched findings from specialized agen-
cies.21 However, it may also consist of newsletter
recommendations from investment banks or other
analysts who base these recommendations on models
that are proprietary knowledge. This means that the
methodologies that produce these findings are impos-
sible to verify, and therefore their objectivity is open
to question.22 Unless investment banks keep research
and trading departments completely independent,
such predictions may well be an attempt to ignite a
new informational cascade and be combined with the
analysts’ prior position-taking, the returns on which
will increase through imitation by others.


If herd behaviour has an impact on price move-
ments, early movers will benefit the most. Imitation
by followers will gradually become less profitable
the longer it is delayed, and the greater becomes the
probability that newly arriving public information
will alter the informational cascade. The speed at
which opportunities for high returns and incentives to
engage in herding behaviour decline, and the extent
to which herding affects prices, depend on the degree
of uncertainty. When it is difficult to differentiate
between uninformed traders, who are herding, and
informed traders, market participants may believe,
mistakenly, that most traders possess accurate infor-
mation. The ensuing confusion allows uninformative
herd behaviour to have dramatic effects on prices
and can lead to bubbles and excessive volatility
(Avery and Zemsky, 1998). Such situations occur
when the prevalence of uninformative noise trading
is underestimated, either because of a lack of data on
the relative importance of different trader categories,
or because of the mistaken belief that trading from
rational arbitrageurs will instantaneously balance




23Financialization of Commodity Price Formation


any price effect from trading that is not based on
fundamentals, as discussed below.


The persistence of price deviations from fun-
damental values caused by herding depends on the
speed and efficiency of arbitrage. An arbitrage op-
portunity presents the possibility of earning a positive
return at no risk. Such a possibility will arise if prices
diverge from fundamental values or across markets
on which identical assets are traded. According to the
EMH, an arbitrageur will detect such an opportunity
immediately, act upon it and thereby make such price
divergences disappear. Given that all these actions are
assumed to happen instantaneously, the notion of un-
limited arbitrage implies the absence of any arbitrage
opportunities. It also implies that irrational position-
taking that would drive prices away from fundamental
values will not make profits, and hence be driven
out of the market. Thus, from an EMH perspective,
speculation must be stabilizing (Friedman, 1953).


However, there is widespread agreement that
there are limits to arbitrage (for a recent survey, see
Gromb and Vayanos, 2010). For example, rational
arbitrageurs may not be able to correct mispricing
either because of risk aversion (de Long et al., 1990a)
or because of capital constraints. Shleifer and Vishny
(1997) argue that arbitrageurs may need to use other
people’s capital. If the market initially moves against
the arbitrageurs, they will need to report intermediate
losses. This will cause the arbitrageurs’ client investors
to withdraw part of their money, so that the arbitrageurs
would need to liquidate their positions at a loss. Given
that arbitrageurs are aware of this possibility, they will
exploit arbitrage possibilities only partially.


What is more, it may not even be optimal for
rational arbitrageurs to counter the position-taking
of irrational investors that follow positive feedback
strategies. Instead, they may want to buy and push up
the price following some initial good news, thereby
providing an incentive for feedback traders to ag-
gressively buy the asset. This reaction by feedback
traders will allow the rational arbitrageurs to sell
their positions at a profit. But in so doing, profitable
arbitrage also contributes to the movement of prices
away from fundamentals and feeds short-term price
bubbles (de Long et al., 1990b).


Bubbles may persist even over a substantial
period of time. This can occur when a bubble bursts
only once a sufficient mass of arbitrageurs have sold
out and rational arbitrageurs know that there will


always remain some agents who are overconfident
or pursue momentum-trading strategies. Rational
arbitrageurs who know perfectly well that the bubble
will eventually burst then need to weigh the risk of
overestimating the remaining number of irrational
traders, which would imply losing all capital gains
by getting out too late, against maximizing profits by
riding the bubble as it continues to grow and exiting
from the market just prior to the crash. New public
information about market fundamentals would allow
rational arbitrageurs to synchronize their exit strate-
gies, and thus make the bubble burst earlier (Abreu
and Brunnermeier, 2003). The same may be true for
disclosure of data that indicate the true number of
remaining irrational traders.23


Taken together, the above discussion shows that
financial investors have a variety of motives, either
rational or irrational, for engaging in trend-following
and momentum trading, as well as for engaging in
arbitrage only to a limited extent. As a result, asset
prices deviate from fundamental values for periods
of time long enough to disturb the normal decision-
making process of consumers and of investors in
fixed capital. This is less visible for commodities
than for currencies. For currencies, where the funda-
mentals are obviously the price differentials, due to
currency speculation exchange rates are driven away
from fundamentals – even in the opposite direction
of fundamentals – for extended periods of time, in
some cases for three to five years.


The discussion also shows that herding can
have sizeable detrimental effects since it reduces
the information content of prices, and because, be-
ing based on only a little information, existing price
levels become very sensitive to seemingly small
shocks. Consequently, commodity prices risk being
subject to speculative bubbles, move far away from
fundamental values and display high volatility.


An empirical assessment of herd behaviour is
notoriously difficult. It is particularly difficult to test
models of informational herding where intentional
herding must be distinguished from spurious herding
(which reflects a common and simultaneous reac-
tion to public announcements). Observing market
transactions and prices cannot reveal the factors
that ultimately determine the decisions of market
participants. This is because actions do not reveal
the kind of private information or signals that agents
receive and that motivate their position-taking. For
commodity markets, this problem is exacerbated by




24 Price Formation in Financialized Commodity Markets: The Role of Information


the fact that data on market transactions are avail-
able only in aggregated form and at relatively long
intervals,24 and it is often difficult to pinpoint what
constitutes fundamentals and how they should be
measured and quantified. This is the case especially
in the presence of a variety of big events that may
change fundamentals gradually but permanently, such
as climate change-related events, peak oil concerns
or increasing demand in emerging markets.


Nonetheless, despite these difficulties, a small
number of studies have attempted to test for herd
behaviour in commodity markets. In principle, trend-
following and momentum trading in commodity
markets can be examined by regressing speculative
position-taking over price changes on previous days.
In addition to unresolved questions as to what trader
categories should appropriately be considered as “spec-
ulators”, daily data on speculative position-taking are
not publicly available. Therefore, using confidential
position data from the CFTC, Irwin and Yoshimaru
(1999), based on data for 1988–1989, and Irwin and
Holt (2005), based on data for 1994, found evidence
for the existence of trend-following or momentum
strategies, but they also found that these had relatively
low price effects. However, the data used in these
studies are dated, and thus cannot reveal the effects
of herding behaviour over the past few years.


A recent study by Gilbert (2010a) uses data for
seven commodities (aluminium, copper, crude oil,


maize, nickel, soybeans and wheat) and looks for
evidence of trend-following behaviour in the pric-
ing process itself. Using monthly data for the period
2000–2009, the study finds a single eight-month bub-
ble for copper (February to October 2006), as well as
one-month bubbles for aluminium (May 2006) and
nickel (April 2007). Using daily data for the period
2006–2008 for crude oil and the three grains, and for
the period 2000–2008 for the non-ferrous metals, the
study finds clear evidence of price bubbles in copper
trading (2004, 2006 and 2008), weak evidence for
crude oil (first half of 2008), nickel (January–March
2007) and soybeans (early 2008), and clear evidence
of the absence of any bubble for aluminium, maize
and wheat. However, Gilbert emphasizes that the
results must be interpreted with caution because
the identification of bubbles may be sensitive to the
selection of the initial date for the sample,25 and also
because explosive price developments may indicate
buoyant fundamentals (i.e. spurious herding) rather
than speculative bubbles.


While the study by Gilbert (2010a), as well as
that by Phillips and Yu (2010), indicate that price
bubbles have developed across commodity markets
over the past 10 years, their results are subject to the
difficulty of separating spurious and intentional herd-
ing, and in particular they do not identify the market
participants that may be responsible for creating and
perpetuating the price bubble.26 The latter issue is
addressed in the next section.


4.5. The price effects of the financialization of commodity markets


The impact of financial traders on commodity
prices is difficult to quantify. Part of this difficulty
is due to the fact that the financialization of com-
modity trading became a major factor roughly at
the same time as demand for physical commodities
from emerging economies started to increase rapidly.
These roughly simultaneous developments make it
difficult to disentangle their relative price impacts.


Accordingly, most empirical assessments of the
impact of financialization on commodity prices have
emphasized either fundamental supply-and-demand


factors or variables that reflect the financialization
of commodity trading. Given that commodity prices
have been influenced by both factors, both these
groups of studies have found a significant impact
on commodity prices of the variables they selected.
Hence, those that attribute most of the development
of commodity prices over the past few years to
fundamental factors (e.g. Sanders and Irwin, 2010),
as well as those that point to an additional impact
from increased financial investment (e.g. Gilbert,
2010b), have been able to provide empirical support
for their point of view. A prominent recent empirical




25Financialization of Commodity Price Formation


study has included both fundamental and financial
variables (Tang and Xiong, 2010). The results of this
analysis refute the contention that growing demand
from emerging economies was the only driver of the
commodity price hike in 2006–2008. They show that
variables reflecting financialization remain significant
even after controlling for fundamental factors. This
finding suggests that the process of financialization
has caused commodity prices to be determined no
longer simply by supply and demand, but also by a
wide range of financial factors and financial investors.
The resulting change in commodity price dynamics is
likely to persist and seriously affect commodity pro-
ducers’ hedging strategies, as well as many countries’
food and energy policies.


A further analysis by Gilbert (2010a) conducts
Granger-causality tests that relate returns on futures
contracts to changes in the positions of index inves-
tors for the period January 2006–March 2009. For
the seven commodities in the sample, the study finds
that changes in index trader positions Granger-cause
price changes for aluminium, copper, crude oil and
maize, while no such impact is detected for soybeans
and wheat. To the extent that changes in index trader
positions are perceived by other traders as conveying
information – similar to the informational cascades
discussed above – these price effects should persist. In
an additional step, Gilbert (2010a) employs a regres-
sion analysis to examine whether the uncovered price
impacts following position changes of index traders are
persistent effects. The results suggest such persistent
effects to be present for copper, crude oil and wheat.


In a final step, Gilbert estimates the price im-
pact of index-based investments by comparing the
actual price developments with those that would have
prevailed had there been no index investments. This
hypothetical price development is estimated based
on the econometric exercises just mentioned. The
evidence indicates that for crude oil prices, index
investors accounted for 3–10 per cent of the price
increases in 2006–2007, but that their impact rose to
20–25 per cent in the first half of 2008 (figure 10).
Their impact on grain prices is estimated to have
been about half that for oil. Gilbert (2010a: 26, 28)
concludes that during the first half of 2008 “index-
based investment generated a bubble in commodity
futures prices” and that overall “it would be incorrect
to argue that high oil, metals and grains prices were
driven by index-based investment but index investors
do appear to have amplified fundamentally-driven
price movements.”


Structural econometric models that incorporate
both the role of current fundamental supply and
demand factors and expectations about the future
development of those factors also indicate the pres-
ence of price bubbles in commodity markets in
2007–2008. Kaufmann et al. (2008) have attempted to
explain oil price developments on the basis of supply
and demand levels, refinery capacity and expectations
which provide an incentive for inventory storage that
bolsters demand.27 Crude oil prices predicted by the
model were fairly close to actual prices until about
mid-2007, when the predicted prices began to grow
rapidly but the actual prices increased even more
rapidly and started to exceed the predicted prices by
a substantial margin, which in the second quarter of
2008 amounted to about 20 per cent (figure 11).28
This result suggests that fundamental supply and
demand factors pushed stocks downwards and prices
upwards starting from 2003, but in 2007–2008 prices
rose above their fundamental levels.


Prometeia (2008) adopts a similar approach in
examining whether the strong increase in oil prices
between mid-2007 and mid-2008 can be explained by
rational pricing behaviour of market participants or
whether it reflects a bubble. The tests cannot reject the
presence of a bubble. Prometeia (2008) interprets the
evidence as pointing to the role of financial investor
activities on commodity futures markets in acceler-
ating and amplifying price movements that in the
medium and long run are driven by fundamentals.29


It has sometimes been argued that the price
impact of index investments detected prior to the
collapse of commodity prices in 2008 is spurious
because similar hikes could be observed for the
prices of commodities that are not included in the
main indexes – the S&P GSCI and the DJ-UBSCI
(ECB, 2008: 18–20). The commodities which ex-
perienced such a price hike but are not part of these
indexes include iron ore, rice and a number of met-
als, such as cadmium and molybdenum. However, a
recent study by Tang and Xiong (2010) shows that
the co-movement between the prices of different
commodities increased after 2003–2004 (i.e. the
beginning of significant position-taking by index
investors on commodity markets), and that for the
commodities included in the major indexes this in-
crease was significantly more pronounced than for
those not included.


All of the empirical evidence discussed so far
relates to the impact of index investments on the




26 Price Formation in Financialized Commodity Markets: The Role of Information


Figure 10


actual Price develoPments and estimated Price develoPments Without index investors,
selected commodities, 2006–2009


Source: Gilbert (2010a).


0


25


50


75


100


125


150


Jan.
2006


Apr.
2006


Jul.
2006


Oct.
2006


Jan.
2007


Apr.
2007


Jul.
2007


Oct.
2007


Jan.
2008


Apr.
2008


Jul.
2008


Oct.
2008


Jan.
2009


Crude oil
($ per barrel)


0


100


200


300


400


500


600


700


800


Jan.
2006


Apr.
2006


Jul.
2006


Oct.
2006


Jan.
2007


Apr.
2007


Jul.
2007


Oct.
2007


Jan.
2008


Apr.
2008


Jul.
2008


Oct.
2008


Jan.
2009


Actual Counterfactual


Maize
(Cents per bushel)


0


200


400


600


800


1 000


1 200


1 400


Jan.
2006


Apr.
2006


Jul.
2006


Oct.
2006


Jan.
2007


Apr.
2007


Jul.
2007


Oct.
2007


Jan.
2008


Apr.
2008


Jul.
2008


Oct.
2008


Jan.
2009


Wheat
(Cents per bushel)


Mar.
2009


Mar.
2009


Mar.
2009




27Financialization of Commodity Price Formation


2007–2008 commodity price spikes. However, as dis-
cussed in sections 4.1 and 4.2, the relative importance
of index investors has declined while that of money
managers has increased. The question therefore arises
as to what price impact these two trader categories
have had over the more recent period. This question
is the focus of the remainder of this section.


Comparing price developments and net finan-
cial positions of different trader categories reveals a
number of salient features (see figure 12 for maize
and figure 13 for crude oil, as well as figure A.1 in the
annex for cocoa, sugar and wheat).30, 31 First, market
participants that have an interest in physical com-
modities (i.e. the category PPMU) almost always take
net short positions (i.e. they are net sellers of futures
and options contracts). Second, financial investors
almost always take net long positions (i.e. they are
net buyers of futures and options contracts). Third,
overall, the comparison provides only scant evidence
of a long-running correlation between index positions
and price changes. While there are clearly periods
and commodities where positions and prices have
moved together, especially during the price collapse


Figure 11


actual and Predicted crude oil Prices,
1997–2008


(Dollars per barrel)


Source: Kaufmann et al. (2008); and private communication from
RK Kaufmann.


0


20


40


60


80


100


1 2 34 1 23 4 12 3 41 2 34 1 23 4 12 3 4 12 3 41 2 34 1 23 4 12 3 41 2 34 1 2


1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 08


NYMEX


Predicted


Figure 12


maize: Prices and net long Financial Positions, By trader category,
June 2006–FeBruary 2011


Source: UNCTAD secretariat calculations, based on weekly data from Bloomberg; and CFTC.
Note: CIT traders = commodity index traders; PMPU = producers, merchants, processors, users.


Maize (Chicago Board of Trade)


-1000


-800


-600


-400


-200


0


200


400


600


13/Jun/2006 02/Jan/2007 01/Jan/2008 06/Jan/2009 05/Jan/2010 04/Jan/2011


N
um


be
r o


f f
ut


ur
es


a
nd


o
pt


io
ns


c
on


tra
ct


s
('0


00
)


0


100


200


300


400


500


600


700


800


C
en


ts
p


er
b


us
he


l


CIT traders (left scale) Price (right scale)PMPU (left scale) Money managers (left scale)




28 Price Formation in Financialized Commodity Markets: The Role of Information


in 2008 and occasionally during the previous price
upturn, there are other times when positions have not
risen during periods of rapid price appreciation. For
example, in the wheat market there was no increase in
either money-manager or index-trader positions dur-
ing the steep price increase from mid-2007 to the end
of the first quarter of 2008. By contrast there appears
to have been a positive correlation between market
positions and maize prices during the same period.
For oil, money-manager positions exhibited strong
volatility, even as oil prices rose almost continu-
ously from the beginning of 2007 through the second
quarter of 2008. Nevertheless, in all the graphs some
correlation between position and price changes is
present over subperiods, as peaks and turning points
seem to occur around the same time.


Fourth, and perhaps most importantly, since
about mid-2009, when commodity prices appear to
have terminated their downward overshooting and
started a relatively stable sideward movement, which
for most commodities ended with the onset of the
price surge in mid-2010, there has been a fairly close
correlation between price changes and changes in
money managers’ positions. This close correlation is


further highlighted by the evidence given in table 3,
where the especially high correlation coefficient for
crude oil is noteworthy.


Figure 13


crude oil: Prices and net long Financial Positions, By trader category,
June 2006–FeBruary 2011


Source: UNCTAD secretariat calculations, based on weekly data from Bloomberg; and CFTC.
Note: CIT traders = commodity index traders; PMPU = producers, merchants, processors, users.


Crude oil (light sweet, NYMEX)


-300


-200


-100


0


100


200


300


13/Jun/2006 02/Jan/2007 01/Jan/2008 06/Jan/2009 05/Jan/2010 04/Jan/2011


N
um


be
r o


f f
ut


ur
es


a
nd


o
pt


io
ns


c
on


tra
ct


s
('0


00
)


0


20


40


60


80


100


120


140


160


$
pe


r b
ar


re
l


Price (right scale)PMPU (left scale) Money managers (left scale)Swap dealers (left scale)


Table 3


simultaneous correlation BetWeen
Price and Position changes, selected
commodities and trader categories,


July 2009–FeBruary 2011
(Correlation coefficient)


Oil Index positions 0.18
Money manager positions 0.81


Cocoa Index positions 0.35
Money manager positions 0.45


Maize Index positions -0.08
Money manager positions 0.52


Sugar Index positions -0.12
Money manager positions 0.54


Wheat Index positions 0.09
Money manager positions 0.56


Source: UNCTAD secretariat calculations, based on data from
Bloomberg; and CFTC.




29Financialization of Commodity Price Formation


Overall, the above evidence indicates that active in-
vestment strategies are increasingly gaining importance
at the expense of the more passive, broad-based index-


investment strategies. It also indicates a close correlation
between commodity prices and the positions of financial
investors that pursue an active trading strategy.


For decades, investments in commodities were
seen as a good opportunity for portfolio diversifica-
tion, because their returns were largely uncorrelated
with those in other markets (see section 4.1). As
explained in the preceding sections, the activities
of financial investors have profoundly affected the
relationship between commodity markets and other
markets over the past decade. Portfolio restructur-
ing, algorithmic trading and herding of market
participants have spilled over from one market to the
other and increased correlations between previously
uncorrelated markets.


With the enhanced financialization of com-
modity markets since the mid-2000s, those markets
have increasingly moved in parallel with financial
markets. During the crisis of 2008, in particular, a
sudden exit of financial players from markets they
considered risky was clearly observable across a wide
range of financialized markets. The first quarter of
2009 marked a turning point, when financial flows
returned once again to risky investments.


Financial investors are usually active in several
financial markets at the same time. Information col-
lected 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 cor-
relations between two asset classes. An increasing
correlation between two markets over time indicates
that the markets have been moving more and more
in tandem.


The evidence from a 30-day rolling correlation
of prices in various asset classes from 1986 to 2010
proves this point. Using box plots32 to capture the


changes in the trend, it was found that the mean cor-
relation fluctuated around zero for most of the period
covered (i.e. from the mid-1980s to the early 1990s).
Significant positive or negative correlations occurred
only in individual years, but not over longer periods.
However, recently this has changed in a number of
markets.


One example is the 30-day rolling correlation
between the WTI front month futures contract and
the Australian dollar–United States dollar exchange
rate shown in the first box plot (figure 14). There is an
upward trend of the median, starting in 2004, that co-
incides with an increasingly concentrated distribution
of correlations. Prior to 2005, the median fluctuated
slightly around a positive value close to zero, but
more recently it has moved regularly beyond 0.5.
It is difficult to construct a substantive relationship
between these variables outside the financial markets
that could explain why the Australian dollar rises
whenever the price of oil increases. With carry trade
for currencies (and the Australian dollar considered
a riskier asset than the United States dollar) and
speculation with oil futures (since commodities are
considered as an alternative investment), the explana-
tion is clear and straightforward.


The same is true for the positive cross-market
correlation between a stock market index like the
S&P 500 and WTI futures (figure 15). Again, the mid-
dle of the last decade marks the beginning of an upward
trend and an increasing correlation. The years 2009
and 2010 show a very compressed box plot, with a
median approaching 1 in 2010. It may be argued that
oil and stocks face rising demand when the global
economy recovers, but the extreme coincidence of oil
and stocks recovering at exactly the same time in the
first quarter of 2009 raises grave doubts about such
a simple argument (see section 4.7 below).


4.6. herding and its effects in different markets




30 Price Formation in Financialized Commodity Markets: The Role of Information


Figure 16 shows the 30-day rolling correlation
between the DJ-UBS Agriculture Total Return Index
and the United States dollar–Brazilian real exchange
rate. There is a persistently high and negative correla-
tion starting in 2007. This means that an appreciation


of the Brazilian real coincides with rising returns on
the DJ-UBS Agriculture Total Return Index. As the
United States dollar–Brazilian real is a preferred
currency pair for carry trade strategies this result
suggests that the same investors are present at the


Figure 14


thirty-day rolling correlation BetWeen the Wti Front month Futures contract and
the australian dollar–united states dollar exchange rate, 1986–2010


Source: UNCTAD secretariat calculations, based on Bloomberg.
Note: For an explanation of the structure of box plots, such as figure 14, see note 32.


1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010


-1


0


0.5


1


-0.5


Figure 15


thirty-day rolling correlation BetWeen the Wti Front month Futures contract and
the s&P 500, 1986–2010


Source: See figure 14.
Note: See figure 14.


1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010


-1


0


0.5


1


-0.5




31Financialization of Commodity Price Formation


same time on both markets, interpreting economic
news in the same fashion irrespectively from indi-
vidual market dynamics (in this case, positively for
risky assets). The relationship between WTI futures
and the United States dollar–Brazilian real exchange


rate is similar. However, the trend towards stronger
correlations begins already in 2005 (figure 17). Since
then, the relationship has become increasingly nega-
tive and persistent.


Figure 16


thirty-day rolling correlation BetWeen the dJ-uBs agriculture total return index and
the united states dollar–Brazilian real exchange rate, 1992–2010


Source: See figure 14.
Note: See figure 14.


-1


0


0.5


1


-0.5


1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010


Figure 17


thirty-day rolling correlation BetWeen the Wti Front month Futures contract and
the united states dollar–Brazilian real exchange rate, 1992–2010


Source: See figure 14.
Note: See figure 14.


-1


0


0.5


1


-0.5


1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010




32 Price Formation in Financialized Commodity Markets: The Role of Information


The results presented above are in line with the
findings of Büyükşahin and Robe (2010: 2, 4; em-
phasis in original): using data not publicly available,
they conclude that “co-movements are positively re-
lated to greater commodity participation by financial
speculators as a whole and by hedge funds especially
– notably by hedge funds that trade in both equity
and commodity futures markets” but that “in contrast
... the positions of other kinds of commodity-futures
market participants (traditional commercial traders,
swap dealers and index traders, floor brokers and
traders, etc.) hold little explanatory power for cross-
market dynamic conditional correlations.”


More recently, cross-market linkages have ap-
peared with a large variety of currencies, stocks and
commodity derivatives, thus reinforcing the evidence
of investor herding in multiple assets. Hedge funds


are widely believed to contribute significantly to
cross-market correlation through the sharing of in-
vestment ideas and by using the same macroeconomic
indicators to formulate their trades.33


Figure 18 illustrates the same relationship in
a different way by using daily prices. It shows the
Brazilian real–Japanese yen exchange rate and se-
lected commodity market indicators for the period
of August 2008 to July 2010 (i.e. which included
a period of sharp swings in all asset prices in 2008
and 2009). Thus a depreciation of the real coincides
with falling commodity prices. The logic is simple: as
commodity futures and emerging-market currencies
are considered risky assets, any flight to security, or
the expectation that the world economy is entering a
period of calm, will move both downwards. More re-
cently, the relationship has been less strong. Since the


Figure 18


relationshiP BetWeen the Brazilian real–JaPanese yen exchange rate and
selected commodity markets, august 2008–July 2010


Source: UNCTAD secretariat calculations, based on Bloomberg.
Note: The vertical axis shows the Brazilian real–Japanese yen exchange rate. Thus, a decrease indicates an appreciation of the


Brazilian real.


R2 = 0.80


0.014


0.016


0.018


0.020


0.022


0.024


0.026


0.028


200 250 300 350 400 450


DJ-UBSCI Commodity Index


E
xc


ha
ng


e
ra


te


R2 = 0.49


0.014


0.016


0.018


0.020


0.022


0.024


0.026


0.028


50 60 70 80 90 100 110 120 130


S&P GSCI Cotton Official Close Index


E
xc


ha
ng


e
ra


te


R2 = 0.41


0.014


0.016


0.018


0.020


0.022


0.024


0.026


0.028


35 40 45 50 55 60 65 70 75 80 85


USDA Cotton Low Middling


E
xc


ha
ng


e
ra


te


R2 = 0.39


0.014


0.016


0.018


0.020


0.022


0.024


0.026


0.028


3,000 4,000 5,000 6,000 7,000 8,000 9,000


S&P GSCI Total Return Index


E
xc


ha
ng


e
ra


te




33Financialization of Commodity Price Formation


summer of 2009, the exchange rate has fluctuated at
around 0.02, whereas commodity prices have contin-
ued to rise. This weakening of the correlation may be
due to the fact that the Brazilian Government became
more and more vocal in criticizing the “unjustified”
appreciation of the real and started to take measures
against carry trades, for example by imposing a tax
on any form of short-term inflows.


To further illustrate the effects of incoming
data on markets where information is processed at
a frequency of one minute or less, we take a much
closer look than the daily observations of the pre-
ceding exercise. The current analysis focuses on the
announcement of United States employment data
on 3 December 2010 as an example (employment
data is considered by Bloomberg as one of “market
moving indicators”). Figure 19 shows the effect of
the publication of the United States labour market
data on the prices of cocoa and WTI futures as well
as the volume of the latter. To obtain a common scale,
prices are rebased to 100 at 13:00 Central European
Time (CET).


On 3 December 2010, the United States employ-
ment data were disappointing. Expectations were
for a stable job market, but instead unemployment
approached the critical level of 10 per cent (9.8 per


cent compared with an expected 9.6 per cent). Within
minutes of the announcement the prices of cocoa and
WTI, among others, dropped sharply, while trade vol-
umes of WTI futures surged. This is not what would
be expected to happen in a market where decisions
are based purely on fundamentals. It is true that the oil
price should have a strong link with economic activ-
ity, and that such activity is reflected in labour market
data. 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 indica-
tor) 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.


These price reactions, therefore, must be due
to the spillover of the financial market logic to com-
modity markets. Financial markets focus strongly
on the release of any economically relevant new
information. A large number of media keep market
participants aware of publication dates, and financial
institutions provide forecasts of economic indica-
tors shortly before their publication. Thus knowing
that “the community” is looking at this information,
market participants form expectations about potential
reactions in the markets, and accordingly, about price


Figure 19


eFFects oF announcement oF emPloyment data in the united states
(reBased series), 3 decemBer 2010


Source: UNCTAD secretariat calculations, based on Bloomberg.


98.5


99.0


99.5


100.0


100.5


101.0


101.5


102.0


13:00 13:15 13:30 13:45 14:00 14:15 14:30 14:45 15:00 15:15 15:30 15:45 16:00 16:15 16:30 16:45 17:00


Pr
ic


e
(re


ba
se


d)


0


500


1,000


1,500


2,000


2,500


3,000


3,500


V
ol


um
e


Employment announcement at
14.30 (CET)


WTI futures price (left scale)
Cocoa futures price (left scale) WTI volume (right scale)




34 Price Formation in Financialized Commodity Markets: The Role of Information


movements. This explains the increased volume of
trading after announcements and the big price effects.
As reactions tend to be similar across a wide range
of markets, the strong cross-market correlations de-
scribed above can be explained easily.


Overall, the evident financialization of com-
modity markets implies that, over significant periods
of time, price changes in the markets do not properly


reflect new information regarding supply and demand
of a specific commodity. The result is an immense
misallocation of resources. Information emanating
from financial markets contaminates the normal price
discovery mechanism in the commodity markets,
thereby generating wrong signals for consumers and
producers, and threatening to make premature adjust-
ment of both extremely costly once the bubbles on
the financial markets burst.


4.7. commodity prices and world business cycles


The most recent decline in world industrial
output is known to have been by far the strongest
of all downward cycles in the past 35 years. The
sharp drop of 12 per cent from the peak makes other
recessions seem like mild slowdowns in comparison
(figure 20). However, in spite of the very low utiliza-
tion of global industrial capacities at the beginning of
2009, the upward pressure on prices in commodity
markets was much stronger than with similar posi-
tions of earlier business cycles – a development often
overlooked by observers. Anticipation of recovery
by the financial markets seems likely to have played
a disproportionately significant role in this current
bout of commodity price inflation.


The strong impact of financial investors on
prices, which may be considered “the new normal of
commodity price determination”, affects the global
business cycle in a very profound way. Commodity
price inflation endangers a smooth recovery to the
extent that it provokes a premature tightening of
monetary policy. It has already played an important
role in the tightening of Chinese and Indian monetary
policy since early 2010, and in the first interest rate
hike since the beginning of the crisis by the European
Central Bank (ECB) in April 2011.


To illustrate this new normal, it is useful to
compare four global business cycles that have oc-
curred since the mid-1970s.34 Global economic
activity may be assumed to be reflected in the
monthly time series of world industrial production
published by the Netherlands Bureau for Economic
Policy Analysis (CPB).35 The periods of recession
troughs can be identified by applying the method


proposed by Bry and Boschan (1971) in BUSY, the
European Commission’s software package. It shows
four recessions for the period 1975–2010, with peaks
in March 1980, October 1981, December 2000 and
March 2008, and respective troughs in September
1980, December 1982, December 2001 and February
2009. To illustrate the cyclical response of financial
markets, the series for industrial production were
normalized by their respective troughs.


A comparison of the business cycles shows
that commodity prices and share prices moved in


Figure 20


dynamics oF World industrial Production
aFter the Peaks oF Four Business cycles


Source: UNCTAD secretariat calculations, based on data from
CPB; and OECD.


a Dating as shown in legend.


88


90


92


94


96


98


100


102


0 2 4 6 8 10 12 14 16 18 20 22 24


Months after peak


In
de


x:
p


ea
k


=
10


0a


March 1980 October 1981
December 2000 March 2008




35Financialization of Commodity Price Formation


opposite directions during the previous identified
business cycles (figures 21–23). By contrast, there
has been a remarkable synchronization of share price
and commodity price movements in the most recent
cycle (figure 24).


This finding supports the results obtained by the
IMF (2010: 31–33) in a similar exercise for devel-
oped economies. In interpreting the results, the IMF
warns against considering the increased synchroniza-
tion of commodity and share prices as evidence in


Figure 21


commodity Prices and market indexes
BeFore and aFter the trough oF


sePtemBer 1980


Source: UNCTAD secretariat calculations, based on data from
ECB; OECD; and UNCTAD.


Figure 22


commodity Prices and market indexes
BeFore and aFter the trough oF


decemBer 1982


Source: See figure 21.


Figure 23


commodity Prices and market indexes
BeFore and aFter the trough oF


decemBer 2001


Source: See figure 21.


Figure 24


commodity Prices and market indexes
BeFore and aFter the trough oF


FeBruary 2009


Source: See figure 21.


0


50


100


150


200


250


300


350


-24 -20 -16 -12 -8 -4 0 4 8 12 16 20 24 28 32 36 40 44 48


Crude oil
S&P 500 Nikkei


Food commodities


No. of months before and after the trough


0


50


100


150


200


250


300


350


-24 -20 -16 -12 -8 -4 0 4 8 12 16 20 24 28 32 36 40 44 48


Crude oil
S&P 500 Nikkei


Food commodities


No. of months before and after the trough


0


50


100


150


200


250


300


350


-24 -20 -16 -12 -8 -4 0 4 8 12 16 20 24 28 32 36 40 44 48


Crude oil
S&P 500 Nikkei


Food commodities


No. of months before and after the trough


0


50


100


150


200


250


300


350


-24 -22-20 -18-16 -14 -12-10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20


Crude oil
S&P 500 Nikkei


Food commodities


No. of months before and after the trough




36 Price Formation in Financialized Commodity Markets: The Role of Information


favour of the financialization of commodity markets,
and affirms that “increased co-movement, however,
likely reflects the sensitivity of both markets to
broader economic developments” (IMF, 2010: 33).
However, such an interpretation neglects to take into
account the low level of capacity utilization in the
wake of the “Great Recession” of 2008 and 2009.
Low capacity utilization, in principle, implies a low
level of industrial use of commodities, and thus a
low level of demand for commodities by their largest
consumers. Under such circumstances, steadily rising
prices of commodities, even ahead of the rebound
of stock market indices, appear to be related more
to an anticipation of a future revival of demand than
to a response to actually rising demand. The most
plausible explanation for such price behaviour is


financialization, which in 2008 eventually led to an
overshooting of commodity prices in both directions
over their “fundamental” levels.


The fact that monetary policy reacts to price
pressure stemming from rising commodity prices,
rather than to bottlenecks in industrial production,
points to a worrisome aspect of the impact of fi-
nancialization that has so far been underestimated,
namely its capacity to inflict damage on the real
economy as a result of sending the wrong signals for
macroeconomic management. This is an important
reason why more effective regulation of commodity
markets is necessary so as to restore an environment
of sound price signals and efficient allocation of re-
sources in today’s modern market economies.




37Field Survey


In addition to the theoretical and empirical
analysis of the way commodity markets function
and how they are affected by the activities of finan-
cial players and a review of the available literature,
interviews were conducted with commodity traders,
financial institutions and other entities which are
closely involved in commodity markets. The inter-
views provide detailed insights into actual market


developments, the process of price formation and
trading strategies as reported by the market partici-
pants themselves. Discussions of regulatory issues
with market participants help gain an understanding
of potential compliance problems or negative side-
effects of regulations. They constitute valuable input
for discussions on current regulations and potential
amendments to them.


5. Field survey


5.1. Objectives


5.2. choice of participants


Interviews were held with various market par-
ticipants in the grain, cocoa, sugar and oil markets.
The emphasis was on physical commodity traders
and financial players such as bank and asset managers
located mainly in Geneva, brokers and consultants
operating in the commodity business were also in-
terviewed (table 4).


The selection of participants followed a multi-
step approach. A list of commodity trading companies
was obtained through Internet research. The Geneva
Trading and Shipping Association (GTSA), a body
representing the interests of Geneva-based companies
engaged in international trade and shipping, also
provided some support in establishing contacts with
traders. The majority of companies did not respond
automatically and had to be contacted several times.
The main reasons for their reluctance to participate in
the study were time constraints, concerns about pro-
viding sensitive information and lack of interest.


From the beginning, the field survey had been
designed as a more qualitative enquiry into current
market practices, and is not intended to be repre-
sentative. As all interviews were on a voluntary basis,
and the researchers had no access to a full list of all


Table 4


classiFication oF intervieWees


Physical traders Grains: 4, crude oil: 3, cocoa: 2,
sugar: 2


Financial players Banks: 5, asset management: 1,
independent financial trader: 1


Others Consulting: 2, brokerage: 1,
price reporting: 1


Note: The numbers refer to the interviews. In some cases
several persons participated in the interview.




38 Price Formation in Financialized Commodity Markets: The Role of Information


relevant entities, a random selection process was
considered unfeasible from the outset. The interview
partners who finally participated represent physical
traders of all the commodities mentioned above. They
work in both small and large companies. Financial
players included bank and asset managers. In addition


one broker, two consultants and representatives of a
price reporting agency were interviewed. As the main
questions concerned the impact of financial investors
on physical markets, the focus was on physical trad-
ers, and they therefore constituted the largest group
interviewed.


5.3. approach


Two separate questionnaires were developed,
one for physical and the other for financial traders.
For crude oil traders one question in the questionnaire
for physical traders was modified (see box 3). The
interviewees received the questionnaire in advance,
which helped them decide whether to participate at
all as well as to prepare for the interview.


The interviews had originally been planned as
personal conversations to take place on the premises
of the participants. However, time constraints and
the preferences of some interviewees made a more
flexible approach necessary. The information was
therefore gathered in three different ways: personal
interviews, telephone interviews and (exceptionally)
written replies to (parts of) the questionnaire(s).


As most of those interviewed asked for strict
confidentiality, it was decided not to record the per-
sonal and telephone interviews. The presence of two
interviewers ensured a better understanding and full
coverage of the interviews. A summary protocol of


each interview was written by one of the interviewers
immediately after the interview and proofread by the
other interviewer. A summary of the findings from the
interviews is reported in this study without disclosing
the identities of the participants or the companies they
work for, as agreed with the participants.


While physical traders and financial traders
were asked to respond to all the questions of the
specific questionnaires, others only needed to reply
to selected questions that were relevant to them. As
all the interviews were voluntary, not all questions
were answered by all the respondents. Some ques-
tions were not answered because the respondent
believed they touched on sensitive business infor-
mation or they simply did not wish to answer. Some
traders did not reply to all the questions owing to
time constraints.


The interviews were conducted between mid-
December 2010 and mid-February 2011 and took
about one hour each on average.




39Field Survey


Box 3


Questionnaire For Physical commodity traders


Introductory and general questions:


1. Where do you conduct your transactions – OTC or on central exchanges? Why do you choose this way
of trading?


For oil traders: Where do you conduct your transactions – OTC or on central exchanges? Why do you
choose this way of trading? What do you trade (WTI or Brent)? Do you prefer one over the other? If
so, why? Do you engage in any arbitrage?


2. Have you felt a growing influence of financial investors in commodity markets and how? Which are
the channels and instruments?


3. Do you consider the influence in “your” commodity market to be particularly strong and, if yes, why?
4. What are the specificities of “your” commodity compared to others?


5. Do you broadly agree with the view that commodity prices in general are more and more determined
by financial investors, instead of by supply and demand? Please explain.


6. What is the impact of physical stocks of commodities on price expectations? How do you assess the
data availability with respect to physical stocks? Do you think that higher international stock holdings
could avoid price spikes?


7. Do you consider current regulations of commodity markets sufficient?


8. Would position limits help to restrict speculation?


9. What other measures would you consider helpful?


10. What is your assessment of the Dodd-Frank Act in this respect?


11. What is your assessment of the European Commission’s regulatory initiative?


12. What problems do you think could be solved by giving a greater role to central counterparties, what
problems would it leave untouched, and what problems could it cause?


13. What else would you consider to be important for a study focusing on the functioning of commodity
markets?


14. What is your personal opinion about the “financialization” and its influence on commodity prices?


Company-specific questions:


1. What kind of information do you rely on for your trading decisions in the short term?


2. Please explain the information flow that you use and the price discovery mechanism on “your” market.
3. Which instruments do you use and why? To what extent do you engage in commodity exchanges, in


OTC-trading, and in instruments like ETFs and ETNs?


4. How do you assess if a commodity is under-valued or over-valued?


5. Does your company try to anticipate position taking by financial investors and how?


6. Have you tried to protect your company’s performance from undue influence of financial investors and,
if so, how?


7. Have you taken purely financial positions or are you thinking of doing so?


8. How have your clients been affected by the increased presence of financial investors in commodity
trading?


9. Has the increased presence of financial investors affected the relationship that your company has with
its clients?


10. Does your company use any sort of algorithm to trade (automatic trading, either High Frequency Trading
(HFT) or simple computer based models)? If yes, what are the main inputs of the model (past prices,
volumes, etc.)?


/...




40 Price Formation in Financialized Commodity Markets: The Role of Information


Questionnaire For Financial traders


Introductory and general questions:


1. In your view, has the influence of financial investors on commodity prices in general increased and how
has it evolved and in which commodity markets has their impact been strongest?


2. What distinguishes trading in “your” commodity from that in others?


3. What is your personal opinion about the financialization of commodities and its influence on prices and
price discovery?


4. Do you agree with the opinion that commodity prices no longer reflect market fundamentals? If yes,
please explain the role of financialization in this.


5. What is the impact of physical stocks of commodities on price expectations? How do you assess the
data availability with respect to physical stocks? Do you think that higher international stock holdings
could avoid price spikes?


6. Could a virtual intervention mechanism as suggested by von Braun and Torero cap speculators’ price
expectations? Or would you consider it easy prey for speculative attacks?


7. Would position-limits help to restrict speculation?


8. What other measures would you consider helpful?


9. What is your assessment of the Dodd-Frank Act in this perspective?


10. What is your assessment of the European Commission’s regulatory initiative?


11. What else do you consider to be important for a study focusing on the functioning of commodity
markets?


Company-specific questions:


1. What kind of information do you rely on for your trading decisions in the short term?


2. Please explain the information flow that you use and the price discovery mechanism on “your”
market.


3. How do you assess if a commodity is under- or over-valued?


4. Which instruments do you use? To what extent do you engage in commodity exchanges, in OTC-trading,
and in instruments like ETFs and ETNs?


5. How has your business been affected by other and new financial investors?


6. How does your company try to anticipate position taking by other financial investors?


7. Has your company bought physical commodities or is it considering doing so?


8. Does your company trade mainly for clients or does it have a proprietary trading desk? If yes, is the
trading strategy similar with the one recommended to clients?


9. Does your company use any sort of algorithm to trade (automatic trading, either High Frequency Trading
(HFT) or simple computer based)? If yes, what are the main inputs of the model (past prices, volumes,
etc.)?


Box 3 (concluded)




41Field Survey


5.4.1. Physical traders


General features of trading activity


The physical traders reported being subject to
strict risk parameters and take only a marginal flat
price exposure if any. Their focus was therefore on
spreads. Only a small minority of traders reported
occasionally speculating on the flat price for a small
share of their overall business.


All grain traders reported trading mainly on
exchanges, CBOT (CME Group) being the most
important, followed by Marché à Terme International
de France (MATIF) (now part of LIFFE). Kansas
City Board of Trade (KCBT) and Minneapolis Grain
Exchange (MGE) were also mentioned, but not by
all traders. Futures appeared to be the most widely
used instrument in grain trade, followed by options,
with swaps being used only to a minor extent. OTC
contracts appeared to be the exception in grain trade.
They help to hedge very specific requirements or bio-
fuel transactions. The respondents reported choosing
exchanges for trading because they are more liquid,
and also because, according to one trader, they are
subject to regulation.


Sugar and cocoa traders also expressed a strong
preference for exchanges, particularly NYMEX, ICE
and LIFFE. The main instrument used is futures, but
also options. OTC contracts seem to be important
where the periods of exchange-traded products do not
match the exposure (e.g. if the delivery month differs
or a longer maturity is required). OTC contracts are
also chosen for longer term hedging.


Trading patterns for crude oil seem to differ
considerably from those for grains and soft commodi-
ties. Exchanges offer only a limited range of products,
which are not sufficient to cover the usual hedging
requirements. Oil traders, in particular, who trade not
only in Brent and WTI, but also in a variety of other
crudes, therefore combine exchange-traded contracts
(WTI, Brent) with more specific OTC contracts to


hedge their exposures. The OTC contracts then serve
to hedge the basis risk. Usually these are swaps (e.g.
WTI-Dubai) which rely on the quotations of a price
reporting agency (e.g. Platts or Argus) that gathers
information on actual prices of different qualities in
different locations on a daily basis.


The use of both European and United States ex-
changes is also related to the time difference between
the two continents.


Sources of information used


Commodity traders reported using a wide range
of information from different sources, the following
being the three major categories of sources:


Official statistics and publicly available re-•
ports (both on “fundamentals” and financial
markets);


Private information obtained from internal com-•
pany sources; and


Communication with other market participants.•


For agricultural commodities, the focus of their
analysis is an assessment of crops and inventories
based on information obtained from a combination
of official statistics, media reports, special reports,
satellite imagery and local or private information.
All grain traders cited monthly data of the USDA
as a vital source of information. In soft commodi-
ties, such as sugar and cocoa, important sources for
statistics are organizations such as the International
Sugar Organization (ISO) or the International Cocoa
Organization (ICCO). In addition, respondents said
they used export and import data by commodity to
assess supply and demand. Larger companies have
their own research departments.


Oil traders mentioned different markets for dif-
ferent kinds of crude, such as light sweet crude oil,
sour crude and heavy crude. For price discovery, oil


5.4. results




42 Price Formation in Financialized Commodity Markets: The Role of Information


firms reportedly rely heavily on the services of price
reporting agencies, though traders of one major oil
company voiced concerns about the reliability of the
reported data.


In addition to generally available information,
traders, especially from large international companies
that cover the whole supply chain, reported using in-
ternal company data (such as inventories at their own
silos, and information on local production in regions
where they are represented). For cocoa, in particular,
local information is needed to assess the crop. Crop
counters move from farm to farm and gather relevant
information. The objective is to have “reliable inside
information well ahead of the markets.”


In addition, most commodity traders mentioned
conversations with other traders – in particular,
former colleagues at other trading companies – and
brokers as an important source of information.
Speaking with peers is considered helpful to obtain
a general idea about the market and for assessing
whether a commodity is over- or undervalued.


For a short-term analysis, real-time market
data provided by Bloomberg or Reuters online are
considered vital. Traders look at open positions and
quantities traded. A number of traders mentioned
technical analysis and other markets, particularly
foreign exchange markets, as being relevant.


The role of fundamentals


There was general agreement among those
surveyed that medium- to long-term price trends are
driven by fundamentals (i.e. supply and demand),
which is why these are the focus of their market
analysis.


Grain traders cited rising demand for feed grains
due to increased meat consumption in emerging
economies as one major demand factor. However,
respondents differed in their assessment of the role
of biofuels: some grain traders believed biofuels to
be a major driver of maize and oilseed prices. Biofuel
production, which requires 40 per cent of United
States maize production and 15 per cent of world
oilseeds production, was mentioned as the reason
why there exists “a tight balance sheet”, which creates
tremendous additional demand for grain. One grain
trader disagreed, arguing that biofuel production is
only one of several factors affecting prices, but not


“a big issue”, as biofuels still account for a limited
share of overall demand.


On the supply side of grain markets, weather
effects and export bans or taxes were mentioned as
uncoupling the local markets from world markets.
It was emphasized that some of the most important
factors are therefore political decisions. Adverse
weather effects (in Australia, Brazil and the Russian
Federation) were also cited as one reason for the cur-
rent spike in sugar prices; another factor is the greater
scarcity of land. Therefore several crops compete for
land and the supply of individual agricultural com-
modities is limited. A respondent cited the example of
cocoa in Malaysia, which has largely been replaced
by palm oil, and some farmers are switching from
cocoa to rubber in Côte d’Ivoire as rubber provides
a steadier income. It was pointed out that the cocoa
supply has grown more slowly than demand over the
past 10 years. Cocoa yields in West Africa are still
relatively low and could be increased substantially
by educating farmers and improving infrastructure.
The then unresolved political crisis in Côte d’Ivoire
was mentioned as the main reason for a “premium”
in the cocoa price in early 2011.


The role of inventories in commodity price
developments was widely acknowledged for all
commodities. It was stressed that data availability
on grain inventories is still limited, especially for
countries like China, whereas sufficiently reliable
data is available for the United States. For cocoa the
stocks-to-usage ratio was considered relatively high,
but so-called “terminal stocks” at the exchanges were
reported to have decreased significantly in Europe.


Speaking in early 2011, oil traders stated that
there is no shortage of oil, and that the oil price is
disconnected from supply and demand.


The role of financial investors


All those interviewed pointed to the impact of
the increasing activities of financial players in com-
modity markets, as evidenced by both rising trading
volumes and increased open interest. It was also noted
that financial players increasingly enter the physical
markets by opening their own trading desks or de-
vising physically backed ETF or ETN. Banks were
also reported to engage in commodity production.
A bank engaged in sugar production for the ethanol
market was cited as an example. Traders differed in




43Field Survey


their perception of time horizons: the majority said
that the strongest impact has been felt over the past
five years or after 2004, while two traders referred
to the past 10 years.


There was a consensus that financial traders
cannot move prices in the long run, but can cause
substantial volatility and price distortions in the short
run. Reasons for their strong short-term price effects
are the enormous volumes of their trades, as well as
the timing of their investments and withdrawals of
funds. Especially in grains and soft commodities,
their relative size is huge in relation to the overall
market. Funds may invest less than 5 per cent of their
money in commodities, but for these relatively small
markets the impact can be enormous. Several traders
said they had the impression that current commodity
prices were higher than they would have been in the
absence of financial traders. Funds, in particular, were
believed to drive up commodity prices.


According to all interviewed traders, financial
investors exacerbate short-run price movements
caused by changes in fundamentals such as supply
disruptions. Further, most financial traders did not
know the specifics of the respective commodity
markets, but based their trade decisions on other
considerations, algorithms or their desired portfolio
structure. Exchange rate developments also spill over
to commodity markets. One grain trader cited the
example of the wheat price on 14 December 2010,
which in his view was not backed by fundamentals
but was due to financial traders aiming to make a
profit at the end of the financial year. Or financial trad-
ers may base their transactions on historical spreads
between wheat and maize, even though these may
no longer be justified by fundamentals.


Volatility makes price discovery more difficult
in all commodity markets. It also makes hedging
more difficult and expensive, as large price move-
ments may trigger margin calls. To meet the margin
calls, traders need sufficient funds of their own or
credit lines from their banks. Several traders men-
tioned difficulties in obtaining such credit lines,
which deter smaller market participants (e.g. farmers)
from hedging at all. According to one cocoa trader,
one important effect of the increased presence of
financial traders is that small trading houses disap-
pear from the market resulting in an increased market
concentration. Respondents noted that “financial
investors have deeper pockets” than physical com-
modity traders. In this context the increasing money


supply due to expansionary policies, particularly in
the United States, was also mentioned. One cocoa
trader mentioned that the functioning of commodity
derivatives markets in providing hedging for physi-
cal commodity traders has become impaired due to
financial investors’ activities.


A grain trader complained that the physical and
the futures markets had moved in different directions
in 2007 and 2008, and that the Chicago exchange
has turned into “a casino” with a number of physical
traders moving away from it because “hedging does
not make sense, when it is riskier to hedge than not
to hedge”. Another grain trader complained that the
“outside money” of financial investors has introduced
a “Wall Street [...] mentality” into the futures markets.
In some cases, divergences between the futures and
the cash markets are also explained by the specifics
of contracts. An example is the cocoa futures contract
at LIFFE, which is in jute bags (bulk trades at a dis-
count), whereas the cash market is in bulk.


High-frequency trading was largely perceived
as problematic; some traders were sceptical about it
because they do not fully understand its impact, while
others criticized the additional volatility caused by
such trading. It was also pointed out that HFT is not
helpful for hedging, because positions are not held
over long time periods. Further, the volatility caused
by HFT discourages hedgers from using the exchang-
es. Traders also voiced concerns about computerized
or algorithm trading. Herding – “They all behave like
lemmings” – was also seen as a problem.


Nevertheless, the overall assessment of finan-
cial players’ presence in commodity markets was
ambiguous. Most traders also saw benefits. They
emphasized that speculators or financial investors36
provide liquidity which is indispensable for hedging.
An oil trader emphasized that one advantage of the
presence of financial players in commodity markets is
the availability of more sophisticated derivatives.


Reaction to financial players’ activities


The majority of the interviewed traders said they
base their trading decisions mainly on fundamentals.
When asked whether they try to anticipate position-
taking by financial traders, however, several replied
that it would make sense to devote more resources
to an analysis of financial players’ activities in com-
modity markets.




44 Price Formation in Financialized Commodity Markets: The Role of Information


Traders of one larger company stressed the
importance of understanding and anticipating fi-
nancial traders’ moves. A large number of analysts
are employed not only to look at fundamentals but
also to anticipate the factors that determine financial
players’ decisions. This approach includes running a
number of models similar to those commonly used
by financial institutions. As one trader put it, “[t]
he banks are trying to understand our markets and
we try to understand their markets.” Consequently,
those traders are increasingly analysing not just the
fundamentals of the commodity markets, but also
financial markets. However, most traders reported
not taking any purely financial positions, while some
stated that they had considered doing so.


Regulations and policy responses


Not all traders were equally well-informed about
derivative market regulations and current reform
efforts. Traders’ interests and views concerning regu-
latory issues varied widely, but most of them agreed
that further regulation is needed, especially in Europe.
Only one oil trader believed that “over-regulation is
a greater danger than under-regulation.”


There was general agreement that more trans-
parency is necessary in commodity markets, and
according to the traders, particularly in Europe, where
hardly any information is available. Most traders felt
that reporting, as by the CFTC, would be a big step
forward. Nevertheless, some of the traders considered
reporting by the CFTC insufficient, with some flaws
in its classification of traders. A number of investment
banks are also physical traders. One grain trader sug-
gested the creation of a “liquidity data bank” to show
who is moving the market on an hourly basis. The
two cocoa traders agreed that United States markets
are sufficiently regulated and transparent, whereas
European markets lack sufficient regulation and
transparency, which encourages market manipula-
tions.37 An oil trader suggested that the CFTC should
ask ICE to provide the same information as NYMEX.
Generally, regulators should know who trades what
both at exchanges and OTC. Data should be made
available after trading. One oil trader believed that
data on positions taken by different groups of trad-
ers should be published daily. The same trader was
sceptical of moving the OTC business to exchanges,
believing that it may actually increase the financiali-
zation of that market segment.


Opinions on position limits were mixed. Many
traders believed that although these are necessary they
are ineffective because they can easily be circum-
vented. For example, positions can be split between
trading platforms or between different subsidiaries of
the same group, and transactions can be carried out
in the OTC market. This is why one trader suggested
there should be limits on OTC transactions as well.
Traders also mentioned the increasing activities of
financial institutions in physical trade. They pointed
out that banks own more and more physical trading
companies, which helps them to obtain hedge ex-
emptions. Concerns were voiced that position limits
might harm hedgers. It was stressed that sometimes a
multiple of the physical value is required “in paper”
to hedge, in which case position limits might hamper
the proper functioning of the market. Position limits
might lead to a situation where particularly physi-
cal traders, for whom the derivatives markets were
established in the first place, would require exemp-
tions all the time. At the same time, investment banks
obtain hedge exemptions because they are involved
in physical trading. Although some traders said that
they would favour stricter position limits, most be-
lieved that their enforcement also has to be improved.
Some traders did not express any opinion on position
limits, one reason being that no information on the
concentration of positions is available.


Generally, most respondents welcomed the
regulatory reforms of the Dodd-Frank Act. However,
one grain trader found the regulations insufficient as
they applied only in the United States, and believed
they should be extended to at least the G-20. The
requirement for OTC transactions to be processed
through clearing houses was seen as one of the big
improvements of the Dodd-Frank Act, and it was
generally agreed as being very necessary. Most trad-
ers seemed at least slightly acquainted with the main
provisions of the Dodd-Frank Act, whereas there was
hardly any awareness of the EU’s regulatory initia-
tive. Therefore the only comment made was that it
was “soft” on commodity markets. Most traders said
they had no comments or would need more informa-
tion before making any comments.


Other suggestions for regulation included intro-
ducing the need to back financial players’ transactions
with more capital (i.e. reduce leverage). One grain
trader called for banning high frequency trading,
since some traders received information earlier than
others, which was unacceptable. A sugar trader sug-
gested that further research be done on the effects




45Field Survey


of high frequency trading before banning it. An oil
trader also suggested that automatic trading be moni-
tored closely. One trader proposed that proprietary
trading of large financial institutions be prohibited.
However, the general attitude towards bans was nega-
tive, as most believed bans to be rather ineffective
and favoured free markets.


Some traders thought the current regulatory
frameworks may be too narrow. While one trader
believed more regulations should apply to the G-20
countries, another thought they should be global
in nature due to economic globalization. Improved
communication between traders and regulators was
also considered urgently necessary, as, in the words
of one trader, “The regulators are lagging behind the
reality.” An oil trader said regulators should be given
more resources.


Grain traders mentioned the lack of convergence
between futures and cash prices as one factor impairing
the functioning of the wheat market. It was also sug-
gested that compelled load-out and additional delivery
locations would be helpful. The former, in particular,
would “kill speculators.” However, the grain traders
did not expect the Chicago Mercantile Exchange to
introduce compelled load-out, because this would
reduce volume, diminishing the CME’s income.


Opinions on strategic inventories were also
divided. Traders who strongly favoured such inven-
tories thought these would help stabilize the markets.
One trader suggested that strategic inventories should
be held by governments in various parts of the world,
and certainly not only by a few investors. However,
another trader emphasized that the market would
work without such intervention. The cost of holding,
especially by individual companies, was mentioned
as an argument against strategic inventories.


5.4.2. Financial players


General features of trading activity


Financial players use all instruments, and trade
both at exchanges and OTC, depending on the needs
of their clients. Financial players also provide their
clients with different types of structured products.
Two of the financial institutions that participated in
the survey concentrate on trade finance and there-
fore partly share the perspective of physical traders.
Generally the financial players interviewed are a less


homogeneous group than the physical traders, and
their experiences and views diverged significantly.


Sources of information used


Financial players mentioned using official sta-
tistics about fundamentals most often. There was a
strong focus on crude oil, where important sources
of information were reported to be data from the
International Energy Agency (IEA) as well as the
Joint Organizations Data Initiative (JODI, see box 1).
Data from the latter are particularly important for
non-OECD countries.


One banker at a large financial institution, who
focuses on the oil market, reported paying much more
attention to financial markets than to fundamentals.
For him the most relevant information includes the
US$ exchange rate, “sentiments in equity and com-
modity markets” and CFTC data. He expressed his
main concern as being, “What is the market thinking?”
For the longer term, GDP growth, the Purchasing
Managers Index (PMI), unemployment data and other
economic indicators are other sources of information.
Nevertheless, he emphasized that financial investors
tend to look at financial data, although they contend
to base their judgement on fundamentals.


The role of fundamentals


Most respondents considered supply and de-
mand as the main drivers of price developments in
the medium term. In the longer term, demography,
strong growth in the BRIC countries (Brazil, the
Russian Federation, India and China) and the effects
of a rising middle class on consumption patterns were
mentioned as demand factors. They envisaged rising
commodity prices owing to insufficient resources.


One banker in trade finance disagreed with the
view that fundamentals determine prices in the medi-
um term. He stated that since 2007–2008 commodity
markets had been disconnected from fundamentals
and that there was a wide gap between the paper
market and the physical market. He doubted that
commodities were in the middle of what he called a
“jumbo cycle”, saying that in reality there was “no
shortage of commodities.” This is why his bank was
also reluctant to finance, for example, the purchase
of physical sugar, as it doubted that the sugar could
be sold at a profit.




46 Price Formation in Financialized Commodity Markets: The Role of Information


Inventories are one variable that clearly has a
strong impact on prices. However, assessments of the
current situation differed widely, also because a lot
of data was viewed as being flawed and or not easily
available (for example, concerning China).


The role of financial investors


The interviewed financial players were less
outspoken about the role of their peers in commod-
ity markets. The general view was that financial
investors have become increasingly interested in
commodities in the past 10 years. One banker, who
focuses on oil, stressed that all commodities have
become an asset class of their own.


Abundant liquidity due to an expansionary
monetary policy and low returns on other assets
were mentioned as one reason for investing in
commodities. The respondents believed that price
effects of financial investors are limited to the short
term. An asset manager said that speculators could
corner the market in the short run because of their
strong financial power. In the short run there appear
to be strong correlations between different markets,
because financial players follow a “risk on/risk off”
pattern in their trading.


The difficulty of gauging the effects of financial-
ization on commodity prices was stressed, because
“speculators” can barely be distinguished from
physical traders. One banker stated that speculators
actually help to mitigate excesses in the markets.


All interviewed financial players agreed that
financial investors could not drive up commodity
prices in the long run.


A striking feature of all interviews was that little
differentiation was made between different types of
financial players, such as index funds or hedge funds.
Many respondents labelled all financial players as
“speculators”.


Regulation and policy responses


As with the physical traders, there was general
agreement that more transparency was a key issue
in commodity markets. For effective regulation it
was necessary to know who was playing in the mar-
ket. Respondents believed Europe needs the sort of


reporting provided by the CFTC in its Commitments
of Traders reports. One banker said it would be
helpful to have more information on the timing of
financial players’ interventions.


A financial trader did not believe financial in-
vestors were an issue. According to him, positions
would not be highly concentrated if there was enough
transparency. With transparency the market would
regulate itself. Therefore OTC transactions should
be regulated and positions published to avoid market
manipulation, position concentration and systemic
risks. Greater transparency of OTC trading was also
mentioned by others as vital, both in Europe and the
United States.


One banker described position limits as “al-
ready very strict”. With regard to position limits,
enforcement seemed the real issue. The respondents
largely agreed that investors would find ways around
position limits, such as splitting positions or starting
physically-backed exchange-traded funds.


They believed that if regulations were too tight,
trading might move to other (less regulated) regions.
One banker contended that the only way to prevent
financial investments from driving commodity prices
would be to stop financial investors from investing in
commodities at all, which would not be feasible.


There were two opposing opinions with regard
to strategic stocks: one group of respondents wel-
comed strategic stocks as a mechanism to stabilize
prices. One person agreed that in principle strate-
gic stocks would be beneficial in the long run, but
building them up would spur further price hikes in
the short run. Others emphasized only the negative
price effects of building stocks and strongly advised
against doing so. Another problem mentioned by
one banker concerned the timing of interventions. If
the market price was seen as a fair price, it would be
difficult to decide whether and when to intervene. It
would also be difficult to assess what would be a fair
price. A virtual intervention mechanism as suggested
by von Braun and Torero (2008) would suffer from
similar flaws.


Several financial players did not express any
particular views on regulation. One banker stressed
that regulating commodity markets would be diffi-
cult, as those markets were highly liberalized and the
players very creative. This is why regulation tends
to be circumvented. He added that the objective of




47Field Survey


regulations should be the creation of a level playing
field for consumers and producers and not that of
obtaining additional taxes.


5.4.3. Others


Interviewees belonging to the third group op-
erate close to the markets. However their business
activities do not usually include position-taking.
One consultant reported that he traded for his own
account at the ICE. Therefore they may be more
neutral market observers.


The role of financial players


There is a general perception that financial
players have played a more and more important role
in commodity markets in recent years. One consult-
ant specializing in oil observed that oil has become
an asset class of its own, whereas a few years ago
it was only traded by informed people. The recent
emergence of exchange-traded funds makes people
trade commodities in the same way as shares. Thus a
large number of market participants do not care about
the fundamentals of the underlying. Consequently,
it is the financial participants who drive prices in
the short run. Although fundamentals always de-
termine the price level in the medium to long term,
short-term fluctuations can be quite damaging to the
market. News agencies that report on the oil market
are also perceived to have an influence. They try
to explain all price movements by fundamentals,
although some price movements are the outcome
of additional inflows of funds into the market or of
technical trading.


One consultant observed that financial invest-
ment in commodities are increasing and will continue
to do so in the coming years. According to him this
has two main effects:


Price volatility increases. There is always some •
volatility in the markets, but financial players
clearly add to it, because their participation
results in more money in the market and often all
market participants move in the same direction.
Over- and undershooting can be as much as
20 per cent.


With financial investors active mostly in deriva-•
tives markets, discrepancies between cash and


futures markets arise that disrupt the markets.
Activities in the cash markets require more exper-
tise and effort. As financial investments increase,
the problems of convergence will get worse.


Another consequence of financial investors’
presence in commodity markets is the increasing
short-run correlation between commodity markets
and other financial markets. If investors face margin
calls in the equity market or some other financial
market, they sell commodity derivatives to meet
margin calls. The ensuing price reactions have no
relation to fundamentals. High-frequency trading
and actively managed funds increase the correlation
between commodity derivatives markets and other
financial markets.


A broker reported noticing a greater involve-
ment of financial players in physical trading and
physical traders in financial markets. According to
him, after a sharp decline of activities (reflected in
outstanding notional amounts), banks that had closed
down their trading desks were now back in the mar-
ket. However, the risk appetite had declined. At the
same time there was more liquidity on the exchanges
than before the crisis, which was being provided by
banks, hedge funds and managed funds. He noted
that commodities were more and more perceived as
“currency” while there was a crisis of fiat money.
The broker also observed increased volatility in
commodity markets due to large inflows of money
from financial investors but also to high leverage
in derivatives markets compared with the physical
market. Not only had the volatility of the flat price
increased, but also that of the spreads.


Regulation and policy responses


The two consultants considered transparency as
a key issue. They stressed that the lack of transpar-
ency made it difficult to regulate commodity markets
without hurting market participants that use deriva-
tives markets purely to hedge. They believed that if
Europeans ensured at least as much transparency as
the CFTC, this would be a big step forward.


One consultant drew attention to the fact that
banks had an information advantage, because they
knew their customers’ trading intentions.


The broker did not express any opinion on trans-
parency, but warned of tendencies to over-regulate in




48 Price Formation in Financialized Commodity Markets: The Role of Information


times of crises. He stated that as commodity markets
were not used by the general public, they could be
regulated to a lesser extent. He also noted that mar-
gins served to limit risk, and saw position limits as
helpful as long as they did not discourage trade.


Political decisions such as the mandate to blend
gasoline with biofuels or subsidizing commodities
were seen as problematic as well as grain subsidies


in the Middle East. Due to the latter, milling grain
(instead of feed grain) was being fed to animals.


One consultant suggested the introduction of
higher initial margins and strict position limits as
steps to mitigate systemic risk which might ensue
from swaps mainly in energy markets. However, he
noted that the industry was already reacting by shift-
ing more business to clearing houses.


There was broad consensus on a number of is-
sues among the commodity market experts who were
interviewed for this study.


The common view was that the role of financial
investors had become more important in recent years.
Due to their financial strength they could move prices
in the short term, leading to increased volatility,
which may harm markets and drive hedgers with an
interest in the physical commodities away from com-
modity derivatives markets. The increased volatility
had resulted in more margin calls and thus higher
financing requirements.


None of the interviewees doubted that com-
modity prices were determined by the fundamentals
of supply and demand in the medium to long term.
Relevant variables were global demand, population
growth, inventories and also political measures, of
which the most frequently mentioned was the promo-
tion of biofuels. Nevertheless, the type of information
used by market participants suggests that financial


market information is much more important for trad-
ing decisions than is commonly acknowledged.


All interviewees agreed that market transpar-
ency needed to be increased, especially in Europe,
where it had important gaps, but also in the OTC
market in the United States. The adoption of report-
ing as provided by the CFTC in its Commitments
of Traders reports would be a big step in the right
direction, but more would be necessary.


Concerning other regulatory issues and aware-
ness of current regulations and reforms, respondents
differed widely. Generally they paid more attention
to United States regulations such as the Dodd-Frank
Act, whereas only a minority of those interviewed
had a clear idea about the European Commission’s
regulatory initiative.


There was substantial scepticism about bans
(e.g. of high-frequency trading) and position limits.
The general belief was that regulations were rather
difficult to enforce.


5.5. summary




49Policy Considerations and Recommendations


The crucial role of information in commodity
price formation has long been recognized, but what
kind of information determines the behaviour of the
most powerful market participants has seldom been
investigated. Is it mainly information about the spe-
cific market of a given commodity or is it information
of a more general nature, i.e. information about the
world economy or about long-term trends that can
hardly be directly related to the existing supply and
demand situation?


In recent years, rapid industrialization, ur-
banization and changes in dietary habits in emerging
economies, especially in Asia, have caused an in-
crease in demand for commodities. And repeated news
about these developments may well have signalled to
market participants the beginning of a new commod-
ity super cycle. This signal from the demand side has
combined with a growing, though at times potentially
deceptive, belief that there are obstacles to a com-
mensurate increase in commodity supply. With regard
to oil, for example, there has been a debate about
whether the point of “peak oil” will be reached in the
near future. With regard to agricultural commodities
(including barley, cocoa, maize, sugar and wheat,
which have been the focus of this study), news about
slower growth of agricultural productivity has added
to already growing concerns about land use, water
shortages, and, more generally, the link between
agricultural production and climate change.


Moreover, first-generation biofuels, which are
based on food stocks, seem to have sharply increased
the relevance of information on energy for trading in
agricultural commodities, and vice versa. The neglect
of investment in research into ways of improving
growth in commodity supply over the past few dec-
ades, when commodity prices were low, is identified as
the main cause of these supply constraints. As a result,


together with uncertainty about demand, a stream of
information on the growing cost of profitable invest-
ment in sustained and resilient commodity supply
growth has signalled to market participants that the
probability of falling commodity prices is rather low.
Consequently, information about fundamental supply
and demand in commodity markets today has been
supplemented by expectations that prices could rise
at any time soon, and for a long period of time.


In such a situation of enhanced price uncertainty,
the traditional roles of commodity futures exchanges
in price discovery and risk management have gained
increasing importance. Commodity exchanges ap-
propriately fulfil this role if market participants, in
addition to using publicly available information, trade
on the basis of independent and individual informa-
tion derived from an intimate knowledge of specific
events relating to commodity markets and on their
own plans to supply or demand commodities.


However, the financialization of commodity
trading has increasingly jeopardized this function of
commodity exchanges. Financial investors in com-
modity markets base their position-taking on risk and
return considerations for which information about
other asset markets and the overall economy play a
key role, as do financial motives more generally. Such
trading behaviour, while relying on similar types of
information, also anticipates the price impact of that
information in similar ways. Taken together, the fi-
nancialization of commodity trading poses the risk of
herd behaviour and of self-fulfilling prophecy due to
the pecuniary power of these market participants.


Even more worrisome is the fact that herding
fundamentally changes the behaviour of markets
and the role that information plays in determining
the right prices. As explained in this study, herding


6. Policy considerations and recommendations




50 Price Formation in Financialized Commodity Markets: The Role of Information


behaviour undermines the specific advantage of free
markets based on the collection of vast amounts of
independent and individual information about supply
and demand. For example, the argument traditionally
presented to defend the participation of powerful fi-
nancial market participants in commodities markets,
namely their role in price discovery and their provi-
sion of liquidity, is no longer tenable once herding
becomes a dominant feature of those markets.


Price discovery by a large number of financial
investors in the market is possible due to the efficient
and quick processing of information about their spe-
cific supply and demand. However, unfortunately it
is usually the wrong price that they “discover”, since
it is not necessarily related to supply and demand in
the market of the specific commodity. Nevertheless,
as this price is easily discovered, it is taken to be the
benchmark price on the market, and often overrules
the prices found in smaller sub-markets or those de-
termined by the price reporting firms such as Platts
and a few others that try to base their pricing on
market fundamentals in the physical markets.38


Similarly, it might be appropriate to question
whether market participants that are subject to


herding actually bring liquidity to the market. A liq-
uid market is one where many different participants
with different sets of information and preferences are
able to find counterparts who are willing to accept
an offer to sell or buy because they have a different
view of how a market is evolving. The Hayekian or
atomistic market mentioned earlier would be char-
acterized by such conditions. A market with a strong
element of herding, which implies that many and
powerful participants use the same information, will
not display those characteristics of differing views
and dispositions.


In light of these developments, it is necessary to
consider how the functioning of commodity futures
exchanges and off-exchange OTC trading could be
improved in a way that would enable commodity
futures exchanges to better fulfil their role of provid-
ing reliable price signals to commodity producers and
consumers, or at least prevent them from sending the
wrong signals. Accordingly, this section examines:
(i) how information and transparency in physical
commodity markets could be improved; (ii) the
need for tighter regulation of financial investors; and
(iii) the need for broader international policy meas-
ures, including price stabilization schemes.


6.1. Improving transparency in physical commodity markets


Greater transparency in physical markets would
enable the provision of more timely and accurate
information about commodities. Comprehensive
information relating to oil would include spare
capacity and global stock holdings, and for agri-
cultural commodities, it would include areas under
plantation, expected harvests, stocks and short-term
demand forecasts. This would allow commercial
market participants to more easily assess current and
future fundamental supply and demand relationships.
Currently, insufficient information makes it difficult
for commercial participants to determine whether
a specific price signal relates to changes in funda-
mentals or to financial market events. This lacuna
also facilitates the intentional introduction of misin-
formation, such as “research-based” price forecasts
by big banks that have taken financial positions in
commodity markets, and can therefore potentially


reap financial benefits if those forecasts turn out to
be accurate. Overall, the availability of high quality
and consolidated, timely information on fundamental
supply and demand relationships in physical markets
would reduce uncertainty and thus the risk of market
participants engaging in herd behaviour.


To achieve greater transparency in physical
markets, there needs to be better producer-consumer
dialogue and improved data collection, analysis and
dissemination. Oil market participants can benefit
from the JODI World Database (see box 1), which
covers production, demand, refinery intake and out-
put, imports, exports, closing stock levels and stock
changes. While this initiative has greatly improved
transparency in the oil market, several gaps remain.
For example, the data are published at monthly
intervals and therefore do not provide adequate




51Policy Considerations and Recommendations


information about short-term events on which active
financial investment strategies are based. Perhaps
more importantly, the database does not include
information on spare capacity. As pointed out by
Kaufmann (2011), it was the lack of information on
spare capacity in non-OPEC oil-producing countries
that caused the sudden slowdown in the growth rate
of non-OPEC crude oil supply after 2004, which
caught market participants by surprise and ignited
a sudden increase in oil prices. Also, the database
does not include information on oil bunkered in
cargo vessels, which is often owned by the private
sector, so that associated information is commercially
sensitive and remains undisclosed. Collecting and
publishing such information in aggregate form in
such a way that its proprietary character would not
be jeopardized would be an important step towards
greater transparency, and could help prevent sharp,
short-term price changes.


There is even less transparency in the physical
market for agricultural products. While information
is available from various sources, as discussed earlier,


the capacity of countries and international organi-
zations to produce consistent, accurate and timely
agricultural market data and analysis remains weak.
Indeed, extreme weather events in both 2007–2008
and 2010 took the international community by sur-
prise. The resulting increased uncertainty may well
have induced misinformed, panic-driven price surges
and triggered increased speculative investment that
amplified the price increases.


Perhaps the most important gap in transparency
in the physical market for agricultural commodities
concerns information on stocks. There are multiple
reasons for poor stock data, a major one being that a
significant proportion of stocks is now held privately,
which makes information on stocks commercially
sensitive. As a result, stock data published by inter-
national organizations are an estimated residual of
data on production, consumption and trade. Enhanced
international cooperation could improve transparency
by ensuring public availability of reliable information
on global stocks. The JODI oil market database could
serve as a model for such an initiative.


The ability of any regulator to understand
what is moving prices and to intervene effectively
depends upon the ability to understand the market
and to collect the required data. However, at present
comprehensive data are not available, particularly
for off-exchange derivatives trading. While traders
on OTC commodity markets benefit from the infor-
mation that traders on organized futures exchanges
provide for price discovery, they do not provide
comparable information of their own.


As expressed in paragraph 13 of the Leaders’
Statement of the G-20 Summit in Pittsburgh in
September 2009, as well as in the conclusions of the
Task Force on Commodity Futures Markets (IOSCO,
2010), which was created by IOSCO in September
2008, transparency on OTC markets could be im-
proved by registering contracts in a trade repository.39
This would be important especially for non-stand-
ardized, illiquid contracts where counterparty risk


involves end users of derivatives who hedge commer-
cial risk in commodities. While such data would need
to remain confidential, their availability to regulators
would reduce the risk of market abuse. The rules pro-
posed by the European Commission (2010), which,
inter alia, envisage central clearing requirements for
standardized contracts, including those involving
index funds, would also help improve transparency
and reduce counterparty risk. In order to capture con-
tracts that are primarily used for speculation rather
than for hedging commodity-related commercial risk,
contracts involving transactions that are intended to
be physically settled should be exempted from such
clearing requirements.40


Significantly more information is available for
trading on commodity futures exchanges, especially
in the United States (as discussed in section 4.2),
which accounts for a substantial proportion of
commodity futures trading. Measures designed to


6.2. Improving transparency in futures and OTC commodity markets




52 Price Formation in Financialized Commodity Markets: The Role of Information


ensure similar reporting requirements for trading on
European exchanges, for which only very limited data


are publicly available at present, would considerably
improve transparency of trading on exchanges.


Regulation of commodity exchanges needs to
find the right balance between being overly restrictive
in the imposition of limits on speculative position
holdings and being overly lax, including in surveil-
lance. If too restrictive, regulation could impair the
hedging functions of commodity exchanges. On the
other hand, if surveillance and regulation are not
strict enough, prices would be able to move away
from levels warranted by fundamental supply and
demand conditions, and would thus equally impair
the hedging functions of the exchanges.


Tighter regulation of financial investors would
facilitate intervention when irregularities are de-
tected. Similar regulations should be adopted in all
commodity exchanges and countries in order to avoid
regulatory migration. In this sense, regulation of the
major commodity exchanges in Europe needs to catch
up with that in the United States, but it also needs
to be tightened in both of them. Tighter regulation
could include the following measures:


An initial measure could be the introduction of •
position limits on individual market participants
and categories of market participants (such as
money managers), as well as on positions of
market participants taken in the same com-
modity but on different exchanges. Exemptions
from such position limits should not be granted
to hedge financial risk, as is the case in the
United States, where swap dealer exemptions
(which also apply to commodity index funds)
are granted with regard to position limits im-
posed on some agricultural commodities. The
issue of position limits is currently under dis-
cussion in both the European Union (European
Commission, 2010) and the United States.41
Such regulatory action relating to positions for


energy commodities, especially those taken by
hedge funds, is also relevant for agricultural
commodities. This is because it has been shown
that hedge funds drive the correlation between
equity and commodity markets, and that food
prices have become more closely tied to energy
prices (Tang and Xiong, 2010; Büyüksahin
and Robe, 2010). However, since the limited
availability of data at present makes it difficult
to determine what levels would be appropri-
ate for position limits, it may take a long time
before such limits can be introduced. As an
interim step, the introduction of position points
could be considered, whereby a trader reaching
a position point would be obliged to provide
further data, on the basis of which regulators
would decide whether or not action is needed
(Chilton, 2011).


A second measure could be the application of •
the Volcker rule (which prohibits banks from
engaging in proprietary trading) to commodity
markets. At present, banks that are involved in
the hedging transactions of their clients have
insider information about commercially based
market sentiment. This enables them to use such
information to bet against their customers. Such
position-taking provides false signals to other
market participants and, given the size of some
of these banks, can move prices away from
levels normally determined by fundamentals,
in addition to provoking price volatility.


A similar rule could be applied to physical traders, •
prohibiting them from taking financial positions
and from betting on outcomes that they are
able to influence due to their strong economic
position in the physical markets (see Blas and
Farchy, 2010, for a recent example).


6.3. Tighter regulation of financial investors




53Policy Considerations and Recommendations


The financialization of commodity futures trad-
ing has made it necessary not only to consider issues
relating to market transparency and regulation, but
also the issue of overcoming excessive commodity
price volatility through supply-side measures. This
is of particular importance for food commodities,
because any sudden increase in demand or major
shortfall in production – or both – when stocks are
low, will rapidly lead to significant price increases.
Physical stocks of food commodities need to be
rebuilt to an adequate level urgently in order to mod-
erate temporary shortages, and to be able to rapidly
provide emergency food supplies for crisis relief to
the most vulnerable.


The accumulation of buffer stocks to smoothen
price volatility and guarantee minimum price levels
has been a controversial issue. It may be difficult
to finance and guarantee the accumulation of suf-
ficiently large physical inventory stocks, especially
of food commodities, for them to function as buffer
stocks. Moreover, it has often been argued that it is
impossible for governments or government agencies
to understand and follow the market. However, in
markets that are driven by herding, any government
agency should be able to understand market devel-
opments to the same extent as market participants
because it has access to similar information as those
participants. 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.


Holding large inventories around the world has
often been judged economically inefficient, and it has
been recommended that net food importing countries
should rely on global markets rather than on building
their own reserves. However, it is clear that newly
imposed trade restrictions (particularly for rice)
played a role in exacerbating the spiralling increase
in food prices in early 2008. This has added to anti-
globalization sentiments and to a more favourable


assessment of the protection that food reserves can
provide.


To counter food price hikes, and as part of
efforts to prevent humanitarian crises, von Braun and
Torero (2008) – echoed by the G-8 summit in June
2008 – have proposed a new, dual global institutional
arrangement: a minimum physical grain reserve for
emergency responses and humanitarian assistance,
and a virtual reserve and intervention mechanism.
The latter would enable intervention in the futures
markets if a “global intelligence unit” were to
deem market prices as differing significantly from
an estimated dynamic price band based on market
fundamentals.


In addition, a multi-tier transaction tax system
for commodity derivatives markets has been pro-
posed. Under this scheme, a progressive transaction
tax surcharge would be levied as soon as prices start
to move beyond the price band defined either on the
basis of commodity market fundamentals (Nissanke,
2010) or on the basis of the observed degree of
correlation between the return on investment in
commodity markets, on the one hand, and equity
and currency markets on the other. Both proposals
deserve due consideration.


Even if such price stabilization mechanisms could
be made to work satisfactorily, it would not make more
physical commodities available on markets, except for
emergency situations. Given that the historically low
level of inventories was one determinant of the abrupt
price hikes in food commodities in early 2008, the
question remains as to what kinds of incentives could
be fostered to increase production and productivity in
developing countries, particularly of food commodi-
ties. Incentives could include a reduction of trade
barriers and domestic support measures in developed
countries. At the same time, increased investment,
including through the provision of more official
development assistance to agriculture in developing
countries, is certainly necessary in this context.




6.4. Price stabilization schemes and other mechanisms






55Conclusions


Financialization has strongly affected the func-
tioning of commodity markets. Due to the increased
participation of financial players in those markets, the
nature of information that drives commodity price
formation has changed. Contrary to the assumptions
of the efficient market hypothesis (EMH), the major-
ity of market participants do not base their trading
decisions purely on the fundamentals of supply and
demand; they also consider aspects which are related
to other markets or to portfolio diversification. This
introduces spurious price signals to the market.


In an environment of substantial uncertainty
with respect to the quality and timeliness of informa-
tion about fundamentals – especially inventories – it
may be a rational strategy to follow others’ strategies
rather than act on one’s own information. This is all
the more so if market participants know that the ma-
jority of their peers are also following such strategies.
The study finds strong evidence confirming such herd
behaviour in commodity markets based on an analysis
of cross-market correlations, price behaviour with
respect to economic announcement and commodity
price behaviour across business cycles. Its findings
support and complement those of other studies.


The analysis was complemented by 22 inter-
views with commodity market participants from
different backgrounds. They reported widespread
herd behaviour and different types of technical trad-
ing. Physical traders, in particular, emphasized that
the activities of financial players have strong effects
on commodity markets and sometimes impair the
functioning of commodity futures for hedging.


In a situation of widespread herding, the as-
sumption of an atomistic market, where participants
trade individually and independently of each other on
the basis of their own interpretation of fundamentals,
thus does not hold any more. The price discovery


mechanism is seriously distorted. Prices can move
far from levels justified by the fundamentals for
extended periods, leading to an increasing risk of
price bubbles. Due to these distortions, commodity
prices do not always provide correct signals about
the relative scarcity of commodities. This impairs
the allocation of resources, has negative effects on
the real economy and leads to food crises, thereby
threatening the lives of the poorest.


To restore the proper functioning of commodity
markets, swift political action is required on a global
scale. It should focus on the following measures:


Increasing transparency in physical markets •
and providing better and more timely data on
fundamentals.


Increasing transparency of OTC markets and •
exchanges by providing more data on market
participants and position-taking, at least to
regulators. This is particularly urgently needed
in Europe.


Tighter regulation, including imposition of •
position limits and banning proprietary trading
by financial institutions that are involved in
hedging the transactions of their clients.


Establishing a government-administered virtual •
reserve mechanism and direct intervention
into the physical or the financial market need
to be considered. In financialized commodity
markets, as in currency markets, intervention
may even make it easier for market participants
to recognize the fundamentals. Moreover,
introducing a transactions tax system which
could generally slow down financial market
activities. All these measures deserve serious
political consideration even if some more
sophisticated schemes may prove difficult to
implement quickly.


7. conclusions






57Annex


Annex table A.1


World Production oF BioFuels


2000 2001 2002 2003 2004 2005 2006 2007 2008 2009


Production (thousands of barrels per day)
World


Total biofuels 315.1 344.3 405.9 501.5 556.3 661.5 854.6 1 127.0 1 489.7 1 635.5
Biodiesel 15.8 21.0 27.5 35.8 44.3 77.2 142.0 202.9 270.9 308.2
Ethanol 299.3 323.3 378.4 465.7 512.0 584.3 712.6 924.1 1 218.8 1 327.3


north america
Total biofuels 109.2 119.6 144.3 187.9 227.3 265.2 340.2 472.8 667.8 767.4
Biodiesel 0.0 0.6 0.7 0.9 1.8 6.1 17.1 33.7 45.9 35.2
Ethanol 109.2 119.1 143.6 186.9 225.5 259.1 323.0 439.2 621.9 732.2


central and south america
Total biofuels 185.1 198.8 221.4 254.8 257.0 285.2 330.6 429.9 539.4 534.4
Biodiesel 0.1 0.2 0.4 0.4 0.5 0.5 2.2 15.2 38.6 57.9
Ethanol 185.0 198.6 221.0 254.4 256.5 284.6 328.3 414.6 500.7 476.4


europe
Total biofuels 17.7 22.6 31.7 41.4 50.4 82.1 141.1 168.6 202.2 234.6
Biodiesel 15.7 20.2 26.3 34.3 41.6 68.1 113.2 137.5 155.0 172.6
Ethanol 2.0 2.4 5.4 7.1 8.8 14.1 27.8 31.1 47.2 62.0


asia and oceania
Total biofuels 2.9 3.1 8.3 17.2 21.1 28.2 41.7 54.2 76.8 93.5
Biodiesel 0.0 0.1 0.1 0.1 0.3 2.2 9.1 15.8 28.8 38.5
Ethanol 2.9 3.0 8.2 17.1 20.8 26.0 32.6 38.4 48.0 54.9


rest of the world
Total biofuels 0.2 0.2 0.2 0.2 0.5 0.8 1.1 1.5 3.5 5.6
Biodiesel 0.0 0.0 -0.0 0.0 0.1 0.3 0.3 0.7 2.5 3.9
Ethanol 0.2 0.2 0.2 0.2 0.4 0.5 0.8 0.8 1.0 1.7


year-on-year growth rate (per cent)
World


Total biofuels 9.3 17.9 23.6 10.9 18.9 29.2 31.9 32.2 9.8
Biodiesel 33.0 30.9 30.2 23.8 74.2 83.9 42.9 33.5 13.7
Ethanol 8.0 17.0 23.1 9.9 14.1 22.0 29.7 31.9 8.9


north america
Total biofuels 9.5 20.6 30.2 21.0 16.7 28.3 39.0 41.2 14.9
Biodiesel 22.2 35.5 96.4 236.3 179.9 96.4 36.4 -23.3
Ethanol 9.0 20.6 30.2 20.6 14.9 24.7 36.0 41.6 17.7


central and south america
Total biofuels 7.4 11.3 15.1 0.9 11.0 15.9 30.0 25.5 -0.9
Biodiesel 100.0 100.0 0.0 25.0 8.7 313.7 577.9 153.3 50.0
Ethanol 7.4 11.2 15.1 0.8 11.0 15.3 26.3 20.8 -4.9


europe
Total biofuels 27.4 40.6 30.6 21.6 63.1 71.8 19.5 19.9 16.0
Biodiesel 28.3 30.5 30.5 21.2 63.6 66.4 21.4 12.8 11.3
Ethanol 20.0 125.0 30.7 23.9 60.7 98.0 11.7 51.6 31.5


asia and oceania
Total biofuels 6.9 168.4 107.2 22.4 33.6 47.9 30.0 41.8 21.6
Biodiesel 20.0 16.7 114.3 633.3 313.6 73.6 82.4 33.7
Ethanol 3.4 173.3 108.5 21.6 25.0 25.4 17.8 25.0 14.4


share in world production (per cent)
north america


Total biofuels 34.7 34.7 35.6 37.5 40.9 40.1 39.8 42.0 44.8 46.9
Biodiesel 0.0 2.7 2.5 2.6 4.1 7.9 12.1 16.6 16.9 11.4
Ethanol 36.5 36.8 38.0 40.1 44.0 44.3 45.3 47.5 51.0 55.2


central and south america
Total biofuels 58.7 57.8 54.5 50.8 46.2 43.1 38.7 38.1 36.2 32.7
Biodiesel 0.6 1.0 1.5 1.1 1.1 0.7 1.6 7.5 14.3 18.8
Ethanol 61.8 61.4 58.4 54.6 50.1 48.7 46.1 44.9 41.1 35.9


europe
Total biofuels 5.6 6.5 7.8 8.3 9.1 12.4 16.5 15.0 13.6 14.3
Biodiesel 99.4 95.9 95.6 95.9 93.9 88.1 79.7 67.8 57.2 56.0
Ethanol 0.7 0.7 1.4 1.5 1.7 2.4 3.9 3.4 3.9 4.7


asia and oceania
Total biofuels 0.9 0.9 2.0 3.4 3.8 4.3 4.9 4.8 5.2 5.7
Biodiesel 0.0 0.5 0.4 0.4 0.7 2.8 6.4 7.8 10.6 12.5
Ethanol 1.0 0.9 2.2 3.7 4.1 4.4 4.6 4.2 3.9 4.1


share in total biofuels (per cent)
World


Biodiesel 5.0 6.1 6.8 7.1 8.0 11.7 16.6 18.0 18.2 18.8
Ethanol 95.0 93.9 93.2 92.9 92.0 88.3 83.4 82.0 81.8 81.2


north america
Biodiesel 0.0 0.5 0.5 0.5 0.8 2.3 5.0 7.1 6.9 4.6
Ethanol 100.0 99.5 99.5 99.5 99.2 97.7 95.0 92.9 93.1 95.4


central and south america
Biodiesel 0.1 0.1 0.2 0.2 0.2 0.2 0.7 3.5 7.2 10.8
Ethanol 99.9 99.9 99.8 99.8 99.8 99.8 99.3 96.5 92.8 89.2


europe
Biodiesel 88.7 89.4 83.0 82.9 82.6 82.9 80.3 81.5 76.7 73.6
Ethanol 11.3 10.6 17.0 17.1 17.4 17.1 19.7 18.5 23.3 26.4


asia and oceania
Biodiesel 0.0 3.2 1.4 0.8 1.4 7.8 21.8 29.2 37.5 41.2
Ethanol 100.0 96.8 98.6 99.2 98.6 92.2 78.2 70.8 62.5 58.8


Source: UNCTAD secretariat calculations, based on EIA.




58 Price Formation in Financialized Commodity Markets: The Role of Information


Annex figure A.1


Prices and net long Financial Positions, By trader category, selected commodities,
June 2006–FeBruary 2011


Source: UNCTAD secretariat calculations, based on weekly data from Bloomberg; and CFTC.
Note: CIT traders = commodity index traders; PMPU = producers, merchants, processors, users.


Wheat (Chicago Board of Trade)


-250


-200


-150


-100


-50


0


50


100


150


200


250


300


13/Jun/2006 02/Jan/2007 01/Jan/2008 06/Jan/2009 05/Jan/2010 04/Jan/2011
0


200


400


600


800


1000


1200


1400


C
en


ts
p


er
b


us
he


l


Cocoa (InterContinental Exchange)


-100


-80


-60


-40


-20


0


20


40


60


80


13/Jun/2006 02/Jan/2007 01/Jan/2008 06/Jan/2009 05/Jan/2010 04/Jan/2011
0


500


1000


1500


2000


2500


3000


3500


4000


$
pe


r m
et


ric
to


n


Sugar (InterContinental Exchange)


-600


-400


-200


0


200


400


600


13/Jun/2006 02/Jan/2007 01/Jan/2008 06/Jan/2009 05/Jan/2010 04/Jan/2011
0


5


10


15


20


25


30


35


40


C
en


ts
p


er
p


ou
nd


N
um


be
r o


f f
ut


ur
es


a
nd


o
pt


io
ns


c
on


tra
ct


s
('0


00
)


CIT traders (left scale) Price (right scale)PMPU (left scale) Money managers (left scale)


N
um


be
r o


f f
ut


ur
es


a
nd


o
pt


io
ns


c
on


tra
ct


s
('0


00
)


N
um


be
r o


f f
ut


ur
es


a
nd


o
pt


io
ns


c
on


tra
ct


s
('0


00
)




59Notes


1 The glossary gives a more detailed definition of the EMH.
2 Under certain assumptions the difference between forward


and futures prices can be neglected (see Hull, 2003: 51–52
and 68–69). In theory, the forward and futures prices
should be identical if interest rates are constant. The
difference is noticeable if interest rates are correlated with
the commodity price. Apart from stochastic and correlated
interest rates, differences in real life can be explained inter
alia by taxes, transaction costs and margin requirements at
the exchanges. Pindyck (1990) estimated these differences
for several commodities and concluded that they were
negligible. Therefore, in this section no difference is made
between the two terms.


3 This simplified formula represents the composition of
the futures price. For the actual calculation of the futures
price there are several possibilities, depending on whether
continuous compounding is applied and whether the
storage cost is a fixed amount or proportional to the value
of the asset (for details see Hull, 2003).


4 Volatility is a measure of price variation from one period
to another. A period of high price volatility is characterized
by a large number of price variations.


5 This is evidenced by the frequently quoted examples of
commodity price bubbles created by financial investors,
including the tulip mania in Holland in the 1630s, the
Mississippi Bubble in France and the South Sea Bubble
in England in the early 1700s (Garber, 1990).


6 Financial innovation has played a facilitating role, as
tracking commodity indexes, such as the Standard and
Poor’s Goldman Sachs Commodity Index (S&P GSCI),
is a relatively new phenomenon. Commodity market
deregulation, such as enacted by the Commodity Futures
Modernization Act (CFMA) of 2000, was a further
facilitating factor, as discussed in UNCTAD 2009b:
76–77.


7 Commodity investment can also take the form of holding
physical stocks, but this is generally considered profitable
only for precious metals, or buying shares in enterprises
that produce commodities. However, the correlation
between a firm’s share price and the price of the underlying
commodity may be low, inter alia because of the additional
layer of management and company risk that may swamp
the underlying commodity risk. A recent example is the
movement of the oil price following the oil spill from a
BP oil platform in the Gulf of Mexico: the oil price rose
while BP’s share price fell.


8 In the S&P GSCI, weights are based on five-year averages
of relative world production quantities; energy products


usually account for about two thirds of the total index.
In the DJ-UBSCI, weights are also based on five-year
averages but rely primarily on the relative amount of
trading activity of a particular commodity; weights are
limited to 15 per cent for individual commodities and to
one third for entire sectors in order to allow for a greater
degree of diversification across commodities.


9 A long position is a market position that obligates the
holder to take delivery (i.e. to buy a commodity), in
contrast with a short position, which obligates the holder to
make delivery (i.e. to sell a commodity). The aggregate of
all long open positions is equal to the aggregate of all short
open positions. For individual traders, net long positions
are total long positions minus total short positions.


10 For explanations of these terms, see section 2 and the
glossary.


11 Notional amount refers to the value of the underlying
commodity. However, traders in derivatives markets do
not own or purchase the underlying commodity, hence
notional value is merely a reference point based on
underlying prices. The limited transparency of data on
OTC markets is underlined by the fact that the category
“other commodities” accounted for roughly 40 per cent
of the total OTC exposure in the late 1990s, but now
constitutes 80–90 per cent.


12 The IMF commodity price index (2005=100) declined
from 203 in the second quarter of 2008 to 99 in the
first quarter of 2009. Over the same period of time, the
UNCTAD non-oil commodity index (2000=100) fell from
294 to 188 and the oil price index from 430 to 157.


13 Part of the following discussion draws on UNCTAD,
2009b.


14 More precisely, among the types of firms engaged in
business activities that can be hedged and therefore
classified as “commercial” by the CFTC are merchants,
manufacturers, producers, and commodity swaps
and derivative dealers. The CFTC classifies as “non-
commercial” all other traders, such as hedge funds, floor
brokers and traders, and non-reporting traders (i.e. those
traders whose positions are below the reporting thresholds
set by an exchange).


15 These 12 commodities are: feeder cattle, live cattle, cocoa,
coffee, cotton, lean hogs, maize, soybeans, soybean oil,
sugar, Chicago wheat and Kansas wheat. The reports
have so far not included similar data for energy and
metals markets because, inter alia, many swap dealers
in metals and energy futures contracts have physical
activities on their own account, which makes it difficult


notes




60 Price Formation in Financialized Commodity Markets: The Role of Information


to separate hedging from speculative activities (CFTC,
2008: 48–49). The CFTC explains that these index trader
data should ultimately be considered as estimates because,
for example, “some traders assigned to the Index Trader
category are engaged in other futures activity that could not
be disaggregated …. Likewise, the Index Trader category
will not include some traders who are engaged in index
trading, but for whom it does not represent a substantial
part of their overall trading activity” (CFTC, Commitments
of Traders, Explanatory Notes, available at: http://
www.cftc.gov/MarketReports/CommitmentsofTraders/
ExplanatoryNotes/index.htm).


16 Asymmetric exposure to leverage is one factor that may
create greater risk. For example, when highly leveraged
financial investors with long positions face margin calls
because they are subject to adverse price movements, they
may not be able to pay this additional margin unless they
liquidate their position. At the same time, commercial
participants with short positions may prefer holding
their positions until expiry and accept physical delivery.
In this situation, a sudden and rapid selling pressure
ignited by financial investors will not be contained by
commercial participants. As a result, prices will move
rapidly and excessively. Leverage may be a particularly
important issue on OTC markets because these markets
are characterized by a high degree of concentration and
much higher leverage ratios than usually observed on
futures exchanges. According to the International Swaps
and Derivatives Association (2010), only about 60 per
cent of commodity derivatives trades are collateralized,
compared with over 90 per cent of credit derivatives trades.
The low level of collateralization of the former implies
higher leverage ratios. Although the relationship between
OTC markets, futures exchanges and spot markets is not
entirely clear, the high concentration and high leverage
ratios in OTC markets pose systemic risks to commodity
markets, and financial stability more generally.


17 Uncertainty in decision-making may be a defining
characteristic of commodity markets. This is because:
(i) medium- and longer-term commodity supply and
demand conditions are subject to considerable uncertainty,
for example because of unknown depletion rates of non-
renewable resources and unknown effects of climate
change on agricultural production; (ii) inventory data,
which provide valuable signals for short-term price
expectations, suffer from significant measurement errors
(Gorton, Hayashi and Rouwenhorst, 2007; Khan, 2009);
and (iii) data on current global commodity supply- and-
demand conditions are published with long time lags and
are frequently revised. Therefore, even well-informed
traders must formulate price expectations on the basis of
partial and uncertain data.


18 Experimental evidence on persistent judgemental errors
in decision-making abounds (see, for example, Ariely,
2010).


19 High-frequency trading (HFT) is a technologically
advanced method of conducting algorithmic trading at
ultra-high speed. Contrary to other types of algorithmic
trading, which focus on price levels and maintain positions
over a period of time, HFT traders attempt to benefit from
price volatility and usually close out their positions by
the end of a trading day. HFT has attracted considerable
attention following allegations that it caused the so-called
“flash crash” on United States equity markets on 6 May
2010. Some observers have also blamed algorithmic


trading for the increase in price volatility on sugar markets
since November 2010 (“High-speed trading blamed for
sugar rises”, Financial Times, 8 February 2011).


20 Similar mechanisms apply when investors follow the
advice of analysts who overweigh public information
and underweigh their own private information in their
messages. Conformity to other analysts’ messages
increases investment in the recommended asset and the
associated return. This, in turn, improves the analysts’
reputations.


21 Casual observation suggests that the release of USDA
reports on livestock and agricultural crops have significant
price effects.


22 Such price predictions can have considerable impact if
they come from a reputed source. For example, Arjun
Murti, a Goldman Sachs analyst, gained considerable fame
between 2004 and 2008 when his successive predictions
of ever higher oil prices appeared to be vindicated by
market developments. According to media reports, other
investors questioned whether Goldman Sachs’ own traders
were benefiting from these predictions, but the bank’s
chief executive denied such accusations (“An oracle of oil
predicts $200-a-barrel crude”, New York Times, 21 May
2008).


23 While this “true number” is necessarily hypothetical,
frequent disclosure of disaggregated data on positions
taken by different trader categories in futures exchanges
and OTC markets could provide valuable information in
this context.


24 Cipriano and Guarino (2010), for example, show that in
equity markets intraday herding can have very significant
price effects.


25 Phillips and Yu (2010) indicate that this problem can be
solved by using an information criterion, rather than the
beginning of the data series, to determine the date of the
first observation.


26 Phillips and Yu (2010), on examining the migration of
price bubbles across equity, bond, currency and commodity
markets (cocoa, coffee, cotton, crude oil, heating oil,
platinum and sugar) since the mid-1990s, find a sequence
of price bubbles, each followed by a financial collapse.
They show that with the eruption of the subprime crisis
in August 2007, financial investment transited from the
United States housing and mortgage markets onto certain
commodity and foreign-exchange markets. Growing
awareness of the serious impact of the financial crisis
on real economic activity both in the United States and
globally caused the general collapse of asset prices in
mid-2008. With respect to commodity prices, their results
point to a price bubble in crude oil between March and
July 2008, in heating oil between March and August 2008,
and in platinum between January and July 2008, while
no price bubbles are detected in cocoa, coffee, cotton and
sugar. This supports the finding of Gilbert (2010a), whose
product sample overlaps with that of Phillips and Yu (2010)
only with respect to crude oil, for which he identifies a
price bubble during the first half of 2008. Phillips and
Yu (2010: 26) explain that early phases of speculative
bubbles are characterized by only small price divergences
from fundamental values, and are therefore statistically
indistinguishable. This may explain why the estimated
date for when the oil price bubble begun is somewhat later
than the observed beginning of the rapid price increase.


27 More precisely, Kaufmann et al. (2008) specify the near-
month price of crude oil on NYMEX as a function of:




61Notes


(i) the equivalent of days of consumption of existing
OECD crude oil stocks; (ii) a factor that reflects OPEC
capacity utilization, OPEC’s share of global oil production
and the extent to which OPEC members cheat on their
quota; (iii) United States refinery utilization rates, which
may be subject to abrupt temporary disturbances during
the hurricane season; and (iv) expectations as reflected
by the difference between the price for the 4-month and
the price for the 1-month futures contract for WTI on
NYMEX. This difference indicates whether the market is
in backwardation or contango, with contango providing
an incentive to build and hold stocks, thereby bolstering
demand and ultimately prices. On the basis of this
relationship, price changes can be estimated with an error
correction model, where first differences of the above
variables as well as the forecasting errors of previous
periods are taken as independent variables.


28 In March 2011, Goldman Sachs estimated the impact of
speculation on the oil price to be about 20 per cent (see,
http://www.cnbc.com/id/42544993).


29 However, other structural models for the oil market ascribe
much of the recent price developments to fundamental
supply and demand factors. These models do not infer
demand shocks from an econometric model, but treat
repeated revisions of forecasts of real income growth
in emerging and advanced economies as a series of
exogenous demand shocks for the global crude oil market
(e.g. Kilian and Hicks, 2009). However, it is hard to believe
that informed oil traders would be repeatedly surprised by
the impact on oil demand of buoyant growth in emerging
economies. Moreover, any such calculation is extremely
sensitive to assumptions about the short-run price elasticity
of supply and demand.


30 For simplicity, these graphs show the net positions of
only three trader categories. All graphs omit the category
“other speculators”. The graphs for the agricultural
products also omit the category “swap dealers”, whose
positions correspond closely to that of the category
“CIT traders”. Given that no data for the category “CIT
traders” are available for crude oil, the respective graph
shows the category “swap dealers”. However, it should
be noted that, contrary to agricultural commodities, for
energy commodities, such as crude oil, the positions
taken by “swap dealers” and “CIT traders” may differ
significantly. This is because swap dealers operating in
agricultural markets undertake only a few transactions
that are not related to index investments. Swap dealers in
energy markets, by contrast, conduct a substantial amount
of such non-index related transactions, which is the very
reason why the CFTC has excluded energy commodities
from its CIT reports. The CFTC (2008) estimates that in
2007–2008, less than half of the long swap dealer positions
in crude oil futures were linked to index fund positions.
This may also explain why swap dealer positions in crude
oil are significantly more volatile that those in agricultural
markets.


31 Comparable data for barley were not available.
32 In descriptive statistics, a box plot (also known as a box-and-


whisker diagram or plot) is a convenient way of graphically
depicting groups of numerical data through their five-


number summaries: the smallest observation (sample
minimum), lower quartile (Q1), median (Q2), upper
quartile (Q3), and largest observation (sample maximum).
A box plot may also indicate which observations, if any,
might be considered outliers. Box plots display differences
between populations without making any assumptions of
the underlying statistical distribution; in other words, they
are non-parametric. The spacing between the different
parts of the box helps indicate the degree of dispersion
(spread) and skewness in the data and also helps identify
outliers. Box and whisker plots are uniform in their use
of the box: the bottom and top of the box are always the
25th and 75th percentile (i.e. the lower and upper quartiles,
respectively), and the band near the middle of the box is
always the 50th percentile (or the median). The ends of
the whiskers (i.e. the lower and upper adjacent values)
represent the lowest datum still within the 1.5 interquartile
range (IQR) of the lower quartile, and the highest datum
still within the 1.5 IQR of the upper quartile.


33 See “Hedge Funds’ Pack Behaviour Magnifies Swings in
Market”, Wall Street Journal (online), 14 January 2011.


34 In the early 1990s, many countries in the world
experienced recessions, but these recessions did not occur
simultaneously. In Germany, for example, the boom after
reunification delayed the cyclical downturn. For this reason
no recession is identified for the world as a whole.


35 Given that these time series begin only in 1991, for the
period 1975–1991 a proxy series was constructed on
the basis of the growth rates of the industrial production
series of the Organisation for Economic Co-operation and
Development (OECD) for all its member States. OECD
industrial production and world industrial production show
fairly similar dynamics in the early 1990s – that is, before
the strong growth of the emerging economies unsettled this
relationship.


36 Not all traders differentiated between individual groups
of non-commercial traders.


37 NYSE LIFFE initiated a trial period for commitments of
traders’ reports starting on 28 September 2010 (NYSE
LIFFE, 2010). This step followed complaints by cocoa
consumers in early July 2010, and threats to shift business
to the ICE.


38 The fact that “price reporting firms” are needed for price
discovery in the physical markets is a clear indication
that these global markets in general are very different
from the kind of atomistic markets that still dominate
the standard economic models. Usually, commodity
markets are not very transparent, and many of them are
segregated regionally to an extent that gives rise to huge
price differentials.


39 For details on how planned rule-making in the United
States is expected to deal with this issue, see Dodd-Frank
Act 2010, sections 727 and 763, as well as Gensler,
2010.


40 Such exemptions are envisaged in the Dodd-Frank
Act 2010, section 721.


41 The CFTC released its draft proposals on 26 January 2011,
accessible at: http://www.cftc.gov/ucm/groups/public/@
lrfederalregister/documents/file/2011-1154a.pdf.






63References


Abreu D and Brunnermeier MK (2003). Bubbles and crashes.
Econometrica, 71(1): 173–204.


Ariely D (2010). Predictably irrational. New York and London:
Harper Perennial.


Avery A and Zemsky P (1998). Multidimensional uncertainty and
herd behaviour in financial markets. American Economic
Review, 88(4): 724–748.


Baffes J and Haniotis T (2010). Placing the 2006/08 Commodity
Price Boom into Perspective. Policy Research Working Pa-
per, No. 5371, The World Bank, 1555-1558, November.


Banerjee A (1992). A simple model of herd behaviour. Quarterly
Journal of Economics, 107(3): 797–818.


Barclays Capital (2010). Commodities Research, December.
Bikhchandani S, Hirshleifer D and Welch I (1992). A theory


of fads, fashion, custom and cultural change as informa-
tional cascades. Journal of Political Economy, 100(5):
992–1026.


Bikhchandani S and Sharma S (2001). Herd behavior in financial
markets. IMF Staff Papers, 47(3): 279–310.


Blas J and Farchy J (2011). Glencore reveals bet on grain price
rise. The Financial Times, 24 April. Available at: http://
www.ft.com/cms/s/0/aea76c56-6ea5-11e0-a13b-00144-
feabdc0.html#axzz1KtsjV7En.


Bry G and Boschan C (1971). Cyclical Analysis of Time Series:
Selected Procedures and Computer Programs. Available
at: http://www.nber.org/books/bry_71-1.


Büyüksahin B and Robe MA (2010). Speculators, Commodities
and Cross-Market Linkages. Mimeo. Available at http://
papers.ssrn.com/sol3/papers.cfm?abstract_id=1707103.


CFTC (2006). Comprehensive Review of the Commitments of
Traders Reporting Program. Federal Register, 71(119):
35627–35632.


CFTC (2008). Staff Report on Commodity Swap Dealers & Index
Traders with Commission Recommendations. Washington,
DC. Available at: http://www.cftc.gov/stellent/groups/
public/@newsroom/documents/file/cftcstaffreporton-
swapdealers09.pdf.


CFTC (2009). About the commitments of traders reports. Avail-
able at: http://www.cftc.gov/marketreports/commitment-
softraders/cot_about.html.


Chang EC (1985). Returns to Speculators and the Theory of Nor-
mal Backwardation. Journal of Finance, 40(1): 193–208.


Chilton B (2011). Statement Regarding Position Limits and
Interim Position Points, 4 January. Available at: http://
www.cftc.gov/PressRoom/SpeechesTestimony/chilton-
statement010411.html.


Cipriani M and Guarino A (2008). Herd behaviour and conta-
gion in financial markets. The B.E. Journal of Theoretical
Economics, 8(1) Article 24.


Cipriani M and Guarino A (2010). Estimating a structural model
of herd behaviour in financial markets. IMF Working Paper
10/288, December.


De Long JB, Shleifer A, Summers LH and Waldmann RJ (1990a).
Noise trader risk in financial markets. Journal of Political
Economy, 98(4): 703–738.


De Long JB, Shleifer A, Summers LH and Waldmann RJ
(1990b). Positive feedback investment strategies and
destabilizing rational speculation. Journal of Finance,
65(2): 379–395.


Devenow A and Welch I (1996). Rational herding in financial eco-
nomics. European Economic Review, 40(3–5): 603–615.


European Central Bank (ECB) (2008). Monthly Bulletin, Frank-
furt, September.


European Central Bank (ECB) (2010). Oil prices – their deter-
minants and impact on euro area inflation and the macr-
oeconomy. Monthly Bulletin, Frankfurt. August.


European Commission (2008). High prices on agricultural
commodity markets: situation and prospects. A review of
causes of high prices and outlook for world agricultural
markets, Brussels, July.


European Commission (2010). Proposal for a Regulation of the
European Parliament and of the Council on OTC deriva-
tives, central counterparties and trade repositories. Docu-
ment COM(2010) 484 final. Brussels, 15 September.


Fajarnes P (2011). An overview of major sources of data and
analyses relating to physical fundamentals in international
commodity markets. UNCTAD Discussion Paper, Geneva,
forthcoming.


FAO (2008). The State of Food and Agriculture 2008. Biofuels:
prospects, risks and opportunities, Rome, Food and Agri-
culture Organization of the United Nations.


FAO (2009). The State of Food and Agriculture 2009. Livestock
in the balance, Rome, Food and Agriculture Organization
of the United Nations.


FAO (2010). Food Outlook. Global Market Analysis. Rome,
Food and Agriculture Organization of the United Nations.
November.


Friedman M (1953). The case for flexible exchange rates. In
M Friedman, Essays in Positive Economics. Chicago:
University of Chicago Press.


Garber PM (1990). Famous first bubbles. Journal of Economic
Perspectives, 4(2): 35–54.


reFerences




64 Price Formation in Financialized Commodity Markets: The Role of Information


Gensler G (2010). Remarks Before ISDA Regional Conference,
16 September. Available at http://www.cftc.gov/Press-
Room/SpeechesTestimony/opagensler-50.html.


Gilbert CL (2010a). Speculative influences on commodity
futures prices 2006–2008. UNCTAD Discussion Paper
No. 197, March.


Gilbert CL (2010b). Commodity speculation and commodity
investment. FAO Commodity Market Review 2009–2010:
26–46.


Gorton G, Hayashi F and Rouwenhorst KG (2007). The fundamen-
tals of commodity futures returns. Working Paper No. 13249.
National Bureau of Economic Research (NBER), July.


Gorton G and Rouwenhorst KG (2006). Facts and fantasies about
commodity futures. Working Paper No. 10595. National
Bureau of Economic Research (NBER), March.


Gromb D and Vayanos D (2010). Limits of arbitrage. Annual
Review of Financial Economics, 2(1): 251–275.


Heap A (2005). China – the engine of a commodities super cycle.
Citygroup Global Markets Inc., Smith Barney.


Helbling T, Mercer-Blackman V and Cheng K (2008). Riding a
Wave. Finance and Development, March: 10–15.


Hernandez M and Torero M (2010). Examining the Dynamic
Relationship between Spot and Future Prices of Agricul-
tural Commodities. IFPRI Discussion Paper, No. 00988,
June.


Hirshleifer D and Teoh S (2003). Herd behaviour and cascad-
ing in capital markets; a review and synthesis. European
Financial Management, 9(1): 25–66.


Hull J (2003). Options, Futures and Other Derivatives, Fifth
Edition. Upper Saddle River, New Jersey.


International Energy Agency (IEA) (2011). Oil Market Report,
Paris.


International Monetary Fund (IMF) (2008). Global Financial
Stability Report, Chapter I, Annex 1.2. Washington, DC.


International Monetary Fund (IMF) (2010). Rebalancing Growth.
World Economic Outlook. Washington, DC.


Intercontinental Exchange (n/d). ICE Crude Oil, Information
Brochure.


International Swaps and Derivatives Association (ISDA) (2010).
Market Review of OTC Derivative Bilateral Collateraliza-
tion Practices, March.


IOSCO (2010). Task Force on Commodity Futures Markets, Re-
port to the G-20, Technical Committee of the International
Organization of Securities Commissions No. OR08/10,
Madrid, November.


Irwin SH and Holt BR (2005). The effect of large hedge fund
and CTA trading on futures market volatility. In: Gregoriou
GN, Karavas VN, Lhabitant FS and Rouah F, eds., Com-
modity Trading Advisors: Risk, Performance Analysis and
Selection, Hoboken (NJ), Wiley.


Irwin SH and Sanders DR (2010). The Impact of Index and
Swap Funds on Commodity Futures Markets: Preliminary
Results. OECD Food, Agriculture and Fisheries, Working
Paper No. 27, Paris.


Irwin SH and Yoshimaru S (1999). Managed futures, positive
feedback trading, and futures price volatility. The Journal
of Futures Markets, 19(7): 759–776.


Kaufmann RK (2011). The role of market fundamentals and
speculation in recent price changes for crude oil. Energy
Policy, 39(1): 105–115.


Kaufmann RK et al. (2008). Oil prices: the role of refinery
utilization, futures markets and non-linearities. Energy
Economics, 30(5): 2609–2622.


Keynes JM (1930). A Treatise on Money, Volume II: The Applied
Theory of Money, London.


Khan MS (2009). The 2008 oil price “bubble”. Policy Brief
09/19. Peterson Institute for International Economics,
Washington DC, August.


Kilian L and Hicks B (2009). Did unexpectedly strong economic
growth cause the oil price shock of 2003–2008? Discussion
Paper No. 7265, Centre for Economic Policy Research,
London, January.


Kolb RW (1992). Is normal backwardation normal? Journal of
Futures Markets, 12(1): 75–91.


Malkiel B (1991). Efficient Market Hypothesis. In: Eatwell J,
Millgate M and Newman P, eds., The New Palgrave: the
world of economics. London and Basingstoke.


McConnell M, Dohlmann E and Haley S (2010). World Sugar
Price Volatility Intensified by Market and Policy Factors.
Amber Waves. United States Department of Agriculture,
Economic Research Service.


Meyer G (2011). Oil price divergence takes crude dealers down
different track. Financial Times 24. 11 February.


Mitchell D (2008). A Note on Rising Food Prices. Policy Re-
search Working Paper, No. 4682. World Bank, July.


NYSE Liffe (2010). Introduction of Enhanced Position Report-
ing Process for Commodity Products and Preparations for
Publication of a “Commitments of Traders” Style Report.
London Info-Flash, No. LO10/23. London, September.


Nissanke M (2010). Mitigating commodity-dependence trap in
LDCs through global facilities. Mimeo. School of Oriental
and African Studies, University of London.


OECD-FAO (2009). Agricultural Outlook 2009–2018, Paris
and Rome.


Phillips PCB and Yu J (2010). Dating the timeline of financial
bubbles during the subprime crisis. Cowles Foundation
Discussion Paper No. 1770, Yale University, September.


Pindyck R (1990). Inventories and the Short-Run Dynamics
of Commodity Prices. MIT-CEPR Working Paper, No.
90–005, March.


Pfuderer S and Del Castillo M (2008). The Impact of Biofuels
on Commodity Prices, Department for Environment, Food
and Rural Affairs, April.


Prometeia (2008). Analisi e previsioni - prezzi delle commodity.
Bologna, October.


Radetzki M (2006). The anatomy of three commodity booms.
Resources Policy, 31(1): 56–64.


Roberts MJ and Schlenker W (2010). Identifying Supply and
Demand Elasticities of Agricultural Commodities: Impli-
cations for the US Ethanol Mandate. National Bureau of
Economic Research Working Paper Series, No. 15921.


Rosegrant M (2008). Biofuels and Grain Prices: Impacts and
Policy Responses. Presented at Testimony for the United
States Senate Committee on Homeland Security and Gov-
ernmental Affairs. Available at: http://www.ifpri.org/sites/
default/files/publications/rosegrant20080507.pdf.


Sanders D and Irwin SH (2010). A speculative bubble in com-
modity futures prices: cross-sectional evidence. Agricul-
tural Economics, 41(1): 25–32.


Scharfstein D and Stein J (1990). Herd behavior and investment.
American Economic Review. 80(3): 465–479.


Scherer B and He L (2008). The diversification benefits of com-
modity futures indexes: a mean-variance spanning test. In:
Fabozzi FJ, Füss R and Kaiser DG, eds., The Handbook of
Commodity Investing. Hoboken (NJ), Wiley: 241–265.


Searchinger T (2008). The impacts of biofuels on greenhouse
gases: How land use alters the equation. Policy Brief of
the German Marshall Fund of the United States.


Shleifer A and Summers LH (1990). The noise trader approach to
finance. Journal of Economic Perspectives, 4(2): 19–33.




65References


Shleifer A and Vishny RW (1997). The limits of arbitrage. Journal
of Finance, 52(2): 737–783.


Steenblik R (2007). BIOFUELS – AT WHAT COST?, Govern-
ment support for ethanol and biodiesel in selected OECD
countries. A synthesis of reports addressing subsidies
for biofuels in Australia, Canada, the European Union,
Switzerland and the United States, Geneva, International
Institute for Sustainable Development, September.


Tang K and Xiong W (2010). Index investment and financiali-
zation of commodities. Princeton University. Working
Paper 16385. National Bureau of Economic Research.
Cambridge (Mass), September.


Trostle R (2008). Global Agricultural Supply and Demand: Fac-
tors Contributing to the Recent Increase in Food Commod-
ity Prices. USDA Report, No. WRS-0801, July.


UNCTAD (2005). Trade and Development Report, 2005 Chapter
II: Income Growth and Shifting Trade Patterns in Asia.
United Nations publication, New York and Geneva.


UNCTAD (2009a). The Biofuels Market: Current Situation
and Alternative Scenarios. New York and Geneva,
UNCTAD.


UNCTAD (2009b). Trade and Development Report 2009, Chap-
ter II: The Financialization of Commodity Markets. United
Nations publication, New York and Geneva.


von Braun J and Torero M (2008). Physical and virtual global
food reserves to protect the poor and prevent market fail-
ure. Policy Brief 4, International Food Policy Research
Institute, Washington, DC, June.






67Glossary


Arbitrage: Transaction that exploits opportunities for risk-
free profits that arise because assets are mispriced.


Algorithmic trading: Trading strategy that bases buying
and selling decisions on computer programmes using
information on past price developments.


Backwardation: Market situation, where futures prices
are progressively lower with rising maturities. (In
some cases the term is used to describe a situation
where the futures price is below the expected future
spot price.)


Carry trade: Carry trade speculation is a strategy in which
an investor sells (e.g. by incurring debt in) a currency
with a relatively low interest rate (i.e. the so-called
“funding currency”) and uses these funds to purchase
short-term assets denominated in a different currency
yielding a higher interest rate.


Contango: Market situation, where futures prices are
progressively higher with rising maturities. (In
some cases the term is used to describe a situation
where the futures price exceeds the expected future
spot price.)


Convenience yield: Utility derived from holding an
inventory.


Cost of carry: Interest and storage cost associated with
inventories.


Derivative: Financial instrument whose value depends on
the value of an underlying asset.


Efficient Market Hypothesis (EMH): An investment
theory that states it is impossible to “beat the market”
because stock market efficiency causes existing share
prices to always incorporate and reflect all relevant
information. According to the EMH, stocks always
trade at their fair value on stock exchanges, mak-
ing it impossible for investors to either purchase
undervalued stocks or sell stocks for inflated prices.
As such, it should be impossible to outperform the
overall market through expert stock selection or


market timing, and that the only way an investor
can possibly obtain higher returns is by purchasing
riskier investments.


Fiat money: Money which only represents a claim on its
issuer, but has no intrinsic value.


High frequency trading is a technologically advanced
method of conducting algorithmic trading at ultra-
high speed.


Index investor: Investor or fund who tracks the move-
ments of an index.


Initial margin: Customers’ funds put up as security for a
guarantee of contract fulfilment at the time a futures
market position is established.


Long position: Position resulting from the purchase of a
derivatives contract.


Noise trader: A trader who bases trading decisions on
considerations which are unrelated to the respec-
tive market thus introducing noise signals into the
market.


Open interest: The total number of futures contracts long
or short in a delivery month or market that has been
entered into and not yet liquidated by an offsetting
transaction or fulfilled by delivery. (Also called open
contracts or open commitments)


Price volatility: A measure of price variation from one
period to another. A period of high price volatility
is characterized by a large number of large price
variations.


Rolling: The process of selling a futures contract before
its expiry and buying a new futures contract of a
later delivery month.


Roll yield: Also called “roll return”, profit from rolling.
Short position: Position resulting from the sale of a de-


rivatives contract.


Swap: An agreement to exchange cash flows in the future
according to a prearranged formula.


Glossary*


* Part of this glossary draws on definitions available on the CFTC’s and the ECB’s websites and on Investopedia.com.






Login