Major International Information Flows Across the Safex Wheat Market

• DOI: http://doi.org/10.1111/saje.12128
• Author: Chris Motengwe,Angel Pardo
• Published date: 01 December 2016
• Date: 01 December 2016

MAJOR INTERNATIONAL INFORMATION FLOWS ACROSS

THE SAFEX WHEAT MARKET

CHRIS MOTENGWE*AND ANGEL PARDO

†

Abstract

We study information flows across four wheat futures markets on four continents: Zhengzhou

Commodity Exchange (ZCE), South African Futures Exchange (SAFEX), Euronext/Liffe and

Kansas City Board of Trade (KCBT). Three approaches for studying information flows among

non-synchronous markets are applied: cointegration techniques, vector autoregressive analysis and

multiple regression proposed. Although comparable underlying assets are traded in the four mar-

kets, our results indicate that no long-run links exist among them. ZCE is by far the most endog-

enous market, and Euronext/Liffe is the most exogenous one. Finally, the model points to KCBT

as the most influential and sensitive wheat market. Our findings indicate that the relative open-

ness of the SAFEX wheat market supports information flows and linkages from KCBTand Euro-

next/Liffe. Therefore, our results suggest that more supportive policies to incentivise higher

wheat production in South Africa are required to mitigate the impact of price shocks emanating

from the global wheat markets.

JEL Classification: C32, G15, G23

Keywords: Wheat futures, information flows, cointegration, vector autoregressive, multiple regression

1. INTRODUCTION

Wheat is one of the most-traded food commodities across international markets and is the

underlying asset both of futures and options contracts in multiple venues.

1

Anumberof

papers have shown the prominent role of US wheat futures markets when contributing to the

international price discovery process (see Garbade and Silber, 1983; Crain and Lee, 1996;

Bessler et al., 2003; G€otz et al., 2013; Hernandez et al., 2014, among others). However,

according to Fung et al. (2010), growing demand in emerging countries and speculation in

* Corresponding author: Wits Business School, University of the Witwatersrand, Johannesburg,

South Africa. Tel: 127832552488, Fax: 127866061998. E-mail: busdevelop7@gmail.com

†

Department of Financial Economics, Faculty of Economics, University of Valencia, Valen-

cia, Spain

The authors are thankful for support by the European Commission under the Erasmus Mun-

dus Action 2 programmes. This funding supported a 6-month mobility in Spain for the first

author under the EU Saturn Project. Special thanks go to Chris Sturgess of the South African

Futures Exchange (SAFEX) for helpful institutional support, suggestions and comments. All

errors remain our own.

1

Compared to other agricultural commodities, wheat globally recorded the largest volumes

traded across international borders during 2014/15. Over this period, USDA (2016) confirms

global wheat imports and exports were 158.3 million tons and 161.3 million tons, respectively,

much higher than corn imports (122.0 million tons) and exports (127.8 million tons) for the same

period.

V

C2016 Economic Society of South Africa. doi: 10.1111/saje.12128

636

South African Journal of Economics Vol. 84:4 December 2016

South African Journal

of Economics

derivative markets that provoked recent food market turmoil in 2006, 2008 and 2010, may

have affected the patterns of information flows across commodity financial markets.

A high number of empirical studies have provided evidence of the dominant role of

futures markets in the price discovery process between spot and futures markets (see Gar-

bade and Silber, 1983; Crain and Lee, 1996; Antonakakis et al., 2015, among others),

surprisingly, as Hua and Chen (2007) indicate, only a few studies have sought to under-

stand the relationship between futures prices of the same underlying asset in different

markets. Within these studies, Geoffrey et al. (1998) analysed the information flows

between US and Canadian wheat futures from 1980 through 1994 and found that

futures prices on Winnipeg Commodities Exchange (WCE) and Chicago Board of Trade

(CBOT) are cointegrated. Balcombe et al. (2007) studied the relationship among maize,

wheat, and soybeans markets in Brazil, USA and Argentina from 1988 through 2001. It

was found that information causality for wheat and soybeans flowed from Argentina and

USA to Brazil. Sendhil and Ramasundaram (2014) analysed wheat information flows

between CBOT and the National Commodities and Derivatives Exchange (NCDEX),

then the largest wheat futures market in India. Following the commencement of wheat

futures trading in India in June 2005, trading in the contract was banned between 2007

and May 2009. In their study, Sendhil and Ramasundaram (2014) investigated informa-

tion flows before and after the banning and no evidence of wheat price cointegration

between CBOT and NCDEX could be confirmed.

Some literature has also focused on agricultural commodity price transmission involv-

ing European Union-based futures markets. Bessler et al. (2003) analysed information

flows in five wheat markets using the error correction method and directed acyclic

graphs. Wheat data from 1981 through 1999 was collected from the Canadian, Austra-

lian, European Union, Argentinian and USA markets. Using monthly free on board

export price quotations for each market, USA wheat prices were found cointegrated with

those of the European Union and Argentina.

Lence et al. (2013) examined long-run linkages between wheat contracts on Chicago

Mercantile Exchange (CME) and Euronext/Liffe. They observed that the CME wheat

futures curve reverts to the mean in the long-term, as opposed to the Euronext curve

which seems not to. Lence et al. (2013) attribute this difference to the fact that CME is

much more liquid than Euronext/Liffe as far as the wheat contracts are concerned. Yang

et al. (2003) examined cross-market linkages of wheat futures in the European Union,

USA and Canada. Data for the study covered 1996 through 2002 and was collected from

London International Financial Futures Exchange (LIFFE), CBOT and WCE. EU prices

were found independent of US prices as opposed to the opposite causal direction where

EU prices significantly influenced US prices in the long-run. Wheat prices in Canada

were found influencing US wheat prices while the reverse relationship was rejected.

The development of Chinese commodity markets has seen increased research focused

on futures contracts behaviour. Du (2004) examined the Zhengzhou Commodity

Exchange (ZCE) wheat market and the CBOT market using data from 1999 to 2003

and found ZCE and CBOT wheat prices not cointegrated. Similar results were obtained

for the same markets by Hua and Chen (2007) using data from 1998 to 2002. Li and

Lu (2012) analysed cross-correlation between USA and Chinese agricultural futures con-

tracts. For small fluctuations, cross-correlations for maize and wheat were persistent in

the short-run. However, cross-correlations for large fluctuations were found not persistent

in the long-run. Finally, Fung et al. (2013) examined 16 futures contracts in China and

637South African Journal of Economics Vol. 84:4 December 2016

V

C2016 Economic Society of South Africa.