Empirical Properties, Information Flow, and Trading Strategies of China's Soybean Crush Spread

• Author: Qingfeng Wilson Liu,Hui He Sono
• Published date: 01 November 2016
• DOI: http://doi.org/10.1002/fut.21777
• Date: 01 November 2016

Empirical Properties, Information Flow,

and Trading Strategies of China’s

Soybean Crush Spread

Qingfeng “Wilson”Liu* and Hui He Sono

This study examines the empirical properties of soybean, soymeal, and soyoil futures prices at

China’s Dalian Commodity Exchange. We find that the three series are cointegrated, and that

the cointegration relationship is characterized by significant seasonality and consistent time

trends. Further, employing a new trivariate VAR-GARCH model, we find evidence of one-way

information flow from the soymeal and soyoil markets to the soybean market, but bidirectional

information flow and volatility spillover between the soymeal and soyoil markets. Trading

simulations based on the mean-reverting tendencies of the cointegration relationship and

5-day averages of the commonly-used spread both generate positive returns. © 2016 Wiley

Periodicals, Inc. Jrl Fut Mark 36:1057–1075, 2016

1. INTRODUCTION

The soybean crush spread measures the profit margin of soybean processors, who buy

soybeans and “crush”them into soybean meal (soymeal) and soybean oil (soyoil).

Soybean processors are interested in using soybean futures crush spread for hedging

purposes, whereas traders seek profitable trading opportunities generated by the spread.

As the world crush capacity increased in recent years, China’s soybean crush capacity

grew rapidly and became the largest soybean crushing country in 2009 (See Figure 1a).

As a result, both China’s soymeal and soyoil production took over the No.1 spot in the

worldandChinabecametheworld’s largest importer of soybeans in 2013 (See

Figure 1b).

1

Fueled by the rapid growth in production and trade, the soymeal, soyoil and soybean

futures contracts at China’s Dalian Commodity Exchange (DCE) also experienced

Qingfeng “Wilson”Liu and Hui He Sono are at the College of Business, James Madison University,

Harrisonburg, VA, USA. The authors would like to thank Alexander Kurov, the editor (Robert Webb), an

anonymous referee, and the participants at the 2014 EFA for comments that helped considerably improve the

paper. This work was supported by the College of Business, James Madison University.

*Correspondence author, MSC 0203, College of Business, James Madison University, Harrisonburg, VA 22807.

Tel: þ(540) 568-3086, Fax: (540) 568-3017, e-mail: liuqx@jmu.edu

Received September 2014; Accepted December 2015

1

Figure 1a and b are from USDA Foreign Agricultural Service. Figure 1a compares the soybean crush capacities of

the top four countries—China, the U.S., Brazil and Argentina from 1964 to 2013. According to the USDA Foreign

Agricultural Service, China’s total supply of soymeal in 2013 was 54.155 million metric tons, and the total supply of

soyoil in China was 14.616 million metric tons. Majority of soymeal and soyoil supply were for domestic

consumption. Figure 1b shows the soybean imports from 1964-2013. China accounts for about 76% of world total

soybean imports in 2013.

The Journal of Futures Markets, Vol. 36, No. 11, 1057–1075 (2016)

© 2016 Wiley Periodicals, Inc.

Published online 14 March 2016 in Wiley Online Library (wileyonlinelibrary.com).

DOI: 10.1002/fut.21777

exceptionally high growth in trading volume from 2008 to 2013.

2

Soybean processors in

China regularly participate in futures trading at the DCE to hedge their price risks. The

hedgers are mostly state-owned enterprises, as well as some private and foreign-funded

processing companies. On the other side of the transaction, the majority of the speculators

are individual market participants, along with some private enterprises as well. Furthermore,

the number of hedgers and speculators has been rising rapidly (Hong, Wang, & Liu, 2009). As

China has become such an important market for soymeal, soyoil, and soybeans, our

investigation of the crush spread in China would be important to the growing community of

futures traders.

A plethora of finance studies have explored the interactions among commodity futures

prices that are related to each other through production processes similar to the crush spread.

Examples include refiners that long gasoline and heating oil and short crude oil futures (the

crack spread), electric utilities that long natural gas and short electricity futures (the spark

FIGURE 1

a: Soybean Crush Capacities (in 1000 Metric Tons, 1964–2013). b: Soybean Imports

(in 1000 Metric Tons, 1964–2013)

Note. Data of Figure 1a and b are from USDA Foreign Agricultural Service: http://apps.fas.usda.gov/

psdonline/psdDownload.aspx.

2

See the annual survey by Will Acworth, the Futures Industry Association, March 2014. According to the survey,

DCE’s soymeal, soyoil and soybean futures contracts were all ranked No. 1 in trading volume in their respective

contracts in the world in 2013, and No. 1, 3, and 19, respectively among all agricultural futures and options

contracts. At the DCE, both soybean and soymeal futures contracts started trading in 1993, while the soyoil contract

was not launched until January 9, 2006.

1058 Liu and Sono