Financialization and commodity prices – an empirical analysis for coffee, cotton, wheat and oil

Abstract

Commodity prices have crucial implications for developing countries. The question whether the financialization of commodity derivative markets has contributed to high and volatile commodity prices has been controversially debated. Building on limitations in the empirical literature, we estimate a multivariate Vector Autoregressive (VAR) model to assess the effect of different groups of financial investors (index investors and money managers) as well as fundamental and macroeconomic variables on the prices of coffee, cotton, wheat and oil. We find that, in contrast to index investors, money managers’ net long positions have a large statistically significant effect on commodity prices. This calls for policy interventions as commodity derivative markets may cease to perform their fundamental developmental roles.

Keywords:

• commodity prices
• commodity derivative markets
• financialization
• Vector Autoregressive models
• coffee
• cotton
• wheat
• oil

JEL classifications:

• C32
• F63
• G15
• O13
• Q11
• Q31

Acknowledgements

The authors would like to thank Bernhard Tröster for research support during the initial phase of this paper, as well as Karin Küblböck, Werner Raza and Georg Lehecka and two anonymous reviewers for very useful comments on a previous draft of the paper. The authors take responsibility for all errors.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Some authors (e.g. Baffes and Haniotis 2010; Gilbert 2010a) however do not find a direct link, or raise doubts that biofuels have accounted for a large shift in global demand.

2. During price increases in the 1970s the stocks-utilization ratio for grains and oilseeds fell to levels as low as 15%; in 2008 the level was only 14% (Carter, Rausser, and Smith 2011).

3. As of July 2008, 11 Asian countries had export controls implemented (Piesse and Thirtle 2009).

4. The distinction into these trader groups follows the classification by the US Commodity Futures Trading Commission (CFTC), which provides weekly market reports (Disaggregated Commitments of Traders (DCOT) reports) on the size of traders’ positions on US commodity futures markets, where financial investors are classified as ‘swap dealers’ and ‘money managers’, as well as a supplemental market report (Commodity Index Trader (CIT) supplement) in which the category of index traders is separately identified.

5. Some index investors also invest in commodity indices by placing their investment directly into futures markets.

6. Related research assesses the effect of financial investors on commodity price volatility and particularly index investors’ positions on commodity futures spreads – the difference between futures prices at different contract maturities. Results are also mixed (for price volatility, see for example Algieri 2012; Bastianin et al. 2012; Bohl, Javed, and Stephan 2012; Brunetti, Büyükşahin, and Harris 2011; Borin and Di Nino 2012; Doroudian and Vercammen 2012; Gilbert 2012; McPhail, Du, and Muhammad 2012; Von Braun and Tadesse 2012 and for futures spreads e.g., Frenk and Turbeville 2011; Hamilton and Wu 2012; Irwin et al. 2011; Stoll and Whaley 2009).

7. Omitted variables can distort results upward and downward depending on the direction of the sign of the correlation between the regressor and the omitted variable. Hence, omitted variables that are positively correlated with the financialization variable, such as global economic activity, would lead to an over-estimation of the impact of financialization.

8. Brent crude oil contains further several missing values. Hence, for some months the average is based on fewer than four or five weeks.

9. The monthly volume of cotton exports is the sum of US cotton Upland and AmPima exports, the neither carded nor combed (NCNC) cotton exports from the EU-27, Brazilian NCNC cotton exports, and total cotton export volumes from India and Australia. For coffee, we obtained monthly total coffee export data from the International Coffee Organization (ICO) upon request. Monthly wheat exports consist of exports from the EU-27, the US and Canada.

10. The sum of demand for and supply of futures, i.e. long and short positions, is always zero. Each trader that sells futures (short positions) needs another trader that buys futures (long positions). Hence, higher total positions or a higher trading volume have no clear effect on prices. What is more interesting is the net positions of certain types of traders. If the number of positions cannot be bought or sold at current prices, it can lead to increasing or declining prices to be able to find a partner for the contract. Clearly, the extent of this pressure also depends on the price-sensitiveness of the respective trading strategies.

11. CFTC classifies: ‘A ‘producer/merchant/processor/user’ as an entity that predominantly engages in the production, processing, packing or handling of a physical commodity and uses the futures markets to manage or hedge risks associated with those activities. A ‘swap dealer’ is an entity that deals primarily in swaps for a commodity and uses the futures markets to manage or hedge the risk associated with those swaps transactions. A ‘money manager’ is a registered commodity trading advisor (CTA); a registered commodity pool operator (CPO); or an unregistered fund identified by CFTC. These traders are engaged in managing and conducting organized futures trading on behalf of clients. Every other reportable trader that is not placed into one of the other three categories is placed into the ‘other reportables’ category (CFTC (US Commodity Futures Trading Commission) 2013a).

12. Net long positions can be either positive or negative. Since taking the logarithm of a negative value is mathematically not possible, this is the only variable that enters the equation untransformed.

13. We applied ADF tests without a trend but with a constant.

14. We used the Final Prediction Error Criterion (FPE), the Akaike Information Criterion (AIC), the Hannan-Quinn Criterion (HQ) and the Schwarz Information Criterion (SIC).

15. This is a reasonable assumption for all commodities except oil, because oil prices usually have an impact on global economic activity. Nevertheless, this impact is unlikely to be observed within one or two months, and can therefore be disregarded.

16. Alternatively, we included the index of global demand for commodities used by Kilian (2009), which does not considerably change our results. The effect remains insignificant.

Additional information

Funding

This work was supported by funds of the Jubiläumsfonds of the Oesterreichische Nationalbank (project number 14686).