Speculators and the futures prices of wheat

On July 2012, the price of wheat took off for the third time in five years. In light of a possible new food emergency, the debate on the role of financial markets has intensified.

The economic literature on the causes of the food price crises can be articulated into two main bodies. On the one hand, there are the empirical studies addressing the issue of fundamental factors and on the other those investigating the speculation hypothesis. Focusing on the latter category, which relates to the specific focus of this paper, two schools of thought dominate. A group of studies finds evidence of a commodity price bubble due to excessive speculation and the other objects this hypothesis.

Within this debate, the investment behavior of participants in futures markets represents an important element of understanding.  Despite this fact, a considerable part of the literature does not accurately distinguish between potential effects caused by speculation in futures markets and the channels of transmission to spot markets.

In this respect, the paper focuses on the speculative activity of different typologies of speculators on the futures market of the Hard Red Winter (HRW) wheat, over the time period from 2000 to 2012, introducing a methodology that makes a distinction between realized effects of futures and spot prices. Potential transmissions of price distortions on derivative to spot markets, caused by speculation, necessarily presume corresponding effects of speculation on futures prices. Therefore, the analysis addresses price formation on futures markets considering the impact on price returns of different non-commercial actors. The speculative variables have been selected and refined according to the data provided by the Commodity Futures Trading Commission (CFTC) and published in different reports.

Concerning the methodology adopted to investigate the impact of speculation on food price, the literature follows different approaches, the Ordinary Least Squares (OLS) regression, the Fama-McBeth cross sectional regression, the Granger causality test and the moving-window Granger causality test. The paper is based on these two latter methods.

It first tests the possible Granger-causal relationship using the full sample. Due to the fact that the literature underlines the possibility of structural breaks in the formation of prices and their relation to speculation on commodity futures markets, the analysis also controls for this possibility introducing  “moving” or  “rolling” windows of data. In this respect the study provides an extension of those by Cooke and Robles (2009) and Robles et al. (2009). However, the paper differs from these two studies and from the majority of the empirical investigations in this field, in the fact that it is not based on year or monthly evidence but on weekly data with a substantial gain in explanatory power.