The untold story of commodity futures in China

• Author: Tingxi Zhang,John Hua Fan
• Date: 01 April 2020
• DOI: http://doi.org/10.1002/fut.22087
• Published date: 01 April 2020

J Futures Markets. 2020;40:671–706. wileyonlinelibrary.com/journal/fut © 2019 Wiley Periodicals, Inc.

|

671

Received: 4 June 2019

|

Accepted: 9 December 2019

DOI: 10.1002/fut.22087

RESEARCH ARTICLE

The untold story of commodity futures in China

John Hua Fan

|

Tingxi Zhang

Department of Accounting, Finance and

Economics, Griffith Business School,

Griffith University, Brisbane, Australia

Correspondence

John Hua Fan, Department of

Accounting, Finance and Economics,

Griffith Business School, Griffith

University, Nathan Campus, 170 Kessels

Road, Nathan, Queensland 4111,

Australia.

Email: j.fan@griffith.edu.au

Abstract

We investigate the behavior of commodity futures risk premia in China. In the

presence of retail‐dominance and barriers‐to‐entry, the term structure and

momentum premia remain persistent, whereas hedging pressure, skewness,

volatility, and liquidity premia are distorted by time‐varying margins and strict

position limits. Furthermore, open interest, currency, and inflation premia are

sensitive to institutional settings. The observed premia cannot be attributed to

common risks, sentiment, transactions costs, or data‐snooping, but are related

to liquidity, anchoring, and regulation‐induced limits‐to‐arbitrage. We highlight

the distinctive features of Chinese futures markets and assess the challenges

posed to theories of commodity risk premia.

KEYWORDS

China, commodity futures, diversification, hedging pressure, liquidity, margins, momentum,

position limits, price discovery, term structure

JEL CLASSIFICATION

G13; G14; G15; G41; N25; Q02

1

|

INTRODUCTION

This paper investigates the behavior of passive long and systematic risk premia in China. Taking a deep dive into the

institutional settings in comparison to the United States, we document the impact of unique market characteristics on

empirical design and results. Our findings contribute to the literature in several ways. First, we conduct a true out‐of‐sample

test of existing theories in a unique environment. These findings inform the pricing of commodities in the presence of retail

dominance, time‐varying margin, and strict position limits. Second, investors seeking new ways to improve their portfolios

have embraced the idea of factor investing globally.

1

In light of the ongoing internationalization process of commodity

markets in China, our findings on systematic risk premia in an emerging futures market are of immediate interests to liquid

alternatives, such as hedge funds and commodity trading advisors who seek for diversification opportunities. Last but not the

least, we discuss policy implications for the efficient functioning of commodity markets.

Since the beginning of the millennium, commodity futures markets in China have undergone tremendous

developments. The trading volume soared from a mere 3 trillion to 227 trillion RMB over the period 2001–2010 (China

Futures Association, 2015).

2

As of 2018, six of the world’s top 10 most active commodity contracts were traded in China.

1

BlackRock (2018) notes the factor industry is $1.9 trillion in assets under management (AUM) and has grown organically at around 11% per year since 2011. In a survey of the world’s 500 largest asset

managers, 5.5% of the AUM of the firms surveyed was invested in factor‐based strategies in 2016 (Willis Towers Watson, 2017). Similarly, Invesco(2017) estimated a 50% growth in proportion of AUM

allocated to factor investing by 2022.

2

A similar trend is observed in the growth of open interest (OI). However, while China dominates the trading volume comparison, the year‐end OIs are consistently larger in the United States. Based

on our selected sample estimates, the dollar or RMB OI is roughly 3.5 times larger in the United States.

Furthermore, China accounted for 56% of global demand for nickel, 50% for coal, copper, and steel, and 47% for

aluminum and pork in 2017 (Visual Capitalist, 2018). To fulfill the economic transition, demand for commodities is

expected to grow even further (FT, 2017). Along with the colossal demand, surging investor interests and the ongoing

liberalization, the once extraneous market in the global commodities trade has become a significant force in

determining the international commodity prices (Bloomberg, 2018b; WSJ, 2016). However, the Chinese commodity

futures market remains enigmatic to practitioners and academics alike due to the barriers‐to‐entry, state influences, and

unique trading rules.

One of the most striking characteristics the literature has become accustomed to, is the large presence of retail

investors. By the end of 2016, more than 86% of the OI was held by individual accounts (China Futures Association,

2016), compared to less than 15% in the United States (Bhardwaj, Gorton, & Rouwenhorst, 2016). Strikingly, the average

holding period of a futures contract in China is less than 4 hours (Citi Group, 2016). Individuals are known to be less

informed (Griffin, Harris, & Topaloglu, 2003), and their trades are more correlated with volumes (Grinblatt &

Keloharju, 2000). Furthermore, retail‐dominance also induces a higher degree of positive feedback trading (De Long,

Shleifer, Summers, & Waldmann, 1990) and herding (Li, Rhee, & Wang, 2017). Therefore, combined with the lack of

strong legal framework (Allen, Qian, & Qian, 2006), a low level of investment risk perception and skills (Wang, Shi, &

Fan, 2006), and evidence of disposition, overconfidence, and representativeness (Chen, Kim, Nofsinger, & Rui, 2007),

the Chinese commodity market is likely highly speculative. The predominant level of sophistication is crucial for the

dynamics of risk premia (Campbell, Grossman, & Wang, 1993).

Moreover, China’s socialist market economy signifies the government’s constant attempt at balancing productivity

and stability. By limiting the market access and maintaining a tight control over its commodity markets, a unique

institutional environment is created for the formation of prices. First, direct participation by foreign investors is

currently restricted, although this is being loosened as part of the effort to internationalize these markets.

3

However,

historically being segregated from global investors likely caused market segmentation, limits‐to‐arbitrage and gradual

information diffusion. For example, Yang, Yang, and Zhou (2012) find that barriers‐to‐entry reduces the information

content of the futures prices in China. Second, directly managed by the government, the priority of the Chinese

exchanges is market stability rather than the maximization of shareholders’wealth. Compared to developed markets

such as the United States, this fundamental difference inevitably leads to more frequent interventions and tighter

regulatory controls, which are ultimately translated into unique implementations of price limits, margins, and position

limits. These unique settings convey significant implications for the futures price discovery and hence the dynamics of

risk premia.

The paper presents four key findings. First, unlike the US markets where nearby contracts are most active, the

nearest to maturity contracts in China account for less than 10% of total volume traded on the futures curve.

Overlooked by the extant literature, this means inferences drawn from the nearest contracts are incomplete at best.

Nonetheless, Ciner (2002) and Webb (1995) find Japanese speculators prefer deferred contracts as they believe there is

more time for their futures position to become profitable. Chinese investors exhibit preferences toward longer‐dated

contracts with specific maturities. Exacerbated by the time‐varying margins and strict position limits, speculators are

involuntarily pushed away from the front end of the futures curve.

4

As a result, a regulation‐induced “limits‐to‐

arbitrage”causes the liquidity and volatility to be lower, and the expected returns to be higher in the nearest to delivery

markets. In these markets, hedgers’ability to effectively transfer risks is impeded.

Second, passive long‐only investing does not yield statistically significant profits, regardless of the time, sector or

weighting schemes employed. This helps explain the relative absence of indexing instruments known to investors.

5

In

general, the performance of long‐only commodities mirrors the rapid growth and recent slowdown of the Chinese

economy. The absence of long‐only premia suggests the theory of normal backwardation (Hicks, 1939; Keynes, 1930)

3

As of October 2019, foreign investors can directly trade in crude oil (from March 2018), iron ore (from May 2018), purified terephthalic acid (PTA) (from November 2018), technically specified rubber

(from August 2019), and fuel oil (from December 2019) markets. Products under discussions include methanol, soybeans, palm olein, linear low‐density polyethylene, polypropylene, copper,

aluminum, nickel, zinc, lead, and tin. These markets are expected to be opened shortly. Since then, qualified foreign brokerage firms are allowed to directly trade in the exchange, instead of conducting

trades via a domestic intermediary.

4

At the time when the paper was written, individual investors in common wheat are prohibited to hold any positions during the delivery month while non individuals are limited to 200 contracts. In

addition, a margin of 5% is applied from the first tothe 15th calendar day of the month before the delivery month. This margin is then increased to 10% for the remaining month and 20% in the delivery

month. A forced liquidation process is triggered if positions exceed the defined limit. As speculators cannot trade sufficient volumes in the nearest contracts, moving to a more distant contract not only

maintains continuous exposures but also avoids a higher margin cost.

5

The Nanhua Commodity Index (www.nanhua.net) covers a broad range of 20 commodities and follows a construction methodology similar to the Commodity Research Bureau. The CFMMC China

Commodity Futures Index (CCFI) was introduced by the China Futures Market Monitoring Center in 2015. The CCFI covers 20 commodities as of January 2018 (www.cfmmc.com/main/views/index.

html).

672

|

FAN AND ZHANG

does not hold well in China. Under this framework, net long speculators accrue a risk premium by assuming the price

risk of net short hedgers. Even though some individual accounts may represent hedgers in disguise, the absence of long‐

only premia suggests that speculators overpower hedgers as a result of retail‐dominance.

Third, unique institutional settings do not eliminate the risk premia. Among an exhaustive list of premia tested, term

structure (TS) and momentum premia are inflated by, but are robust to, strict position limits and time‐varying margins.

As roll‐yields and past returns are empirically linked to inventory (Gorton, Hayashi, & Rouwenhorst, 2013), our

findings suggest that unique trading rules do not disrupt the pricing fundamentals underpinned by the theory of storage

(Working, 1949). However, hedging pressure (HP), volatility, liquidity, and skewness premia can be distorted. Thus, we

argue that the HP hypothesis (Hirshleifer, 1989) and the selective hedging hypothesis (Fernandez‐Perez, Frijns, Fuertes,

& Miffre, 2018) are potentially incomplete. On the other hand, we find the informational content of gross OI,

correlations with currency and inflation, are sensitive to excessive speculation, government intervention and economic

structures, respectively.

6

Therefore, future research on commodity pricing ought to incorporate the effects of unique

institutional settings.

Fourth, the TS and momentum premia in China cannot be attributed to aggregate market risks, non‐tradable

macroeconomic risks or changes in market sentiments. However, illiquidity, anchoring bias, and the regulation induced

limits‐to‐arbitrage provide at least a partial explanation. Moreover, the HP premium can be adequately explained by the

Bakshi, Gao, and Rossi (2019) model. While the broad commodity market in China is more intertwined with stocks

than in the United States, TS, momentum, and HP premia are found to offer promising diversification potential for

Chinese equity exposures, though the benefit is immediate only to Chinese investors. Finally, our evidence suggests

Chinese commodity prices are an important indicator of future economic growth in China. An extensive suite of

robustness tests suggests our findings are consistent in the most liquid markets, randomly excluded commodity sectors,

US dollar‐denominated returns, a “matched”sample with the United States, by altering futures return measurements,

portfolio breakpoints, and cannot be subsumed by data‐snooping or transaction costs.

2

|

INSTITUTIONAL SETTINGS

2.1

|

A brief walk down the great wall of commodity

To better understand the dynamics of commodity futures in China, it is worthwhile to first understand its origin. The

history began with a grain wholesale market established in October 1990 in Zhengzhou, the capital city of Henan

Province, a major agricultural production region. Initially, the grain wholesale market only facilitated spot transactions,

but quickly evolved to futures trading (Williams, Peck, Park, & Rozelle, 1998). Due to the speculative nature of

inexperienced participants and the ineffective regulatory oversight, the market experienced a chaotic period in its first

decade of development. After several boom and bust cycles, a market overhaul took place in 2000. The number of both

the exchanges and products reduced from a peak of more than 50, to just three exchanges and nine products. A

relatively steady period of development followed until the introduction of the index futures in 2010. The financial

futures sparked enormous interests in futures trading, where the commodity futures markets also benefited from the

soaring volume.

However, the regulators quickly introduced several policies to curb the excessive speculation to prevent a potential

collapse. For example, starting from 2011, newly launched contracts were mainly big‐scale.

7

This fundamentally

constrained speculation, and also made the Chinese market more compatible with international counterparts.

Increased margins, price limits, and transaction fees caused the trading lots to drop substantially (by 51%) in 2011 (see

Figure 1). However, the effectiveness of such actions on speculation remains unanswered, as recent studies find higher

margins can lead to a reduction in open positions and increase in volatility (Hedegaard, 2014), whereas price limits may

restrict price discovery but not stop speculation (Janardanan, Qiao, & Rouwenhorst, 2019). After decades of

consolidation, the Shanghai Futures Exchange (SHFE), Zhengzhou Commodity Exchange (ZCE), Dalian Commodity

6

Hong and Yogo (2012) posit the gross OI is a better predictor of future economic activity and asset prices than futures prices. In their model of limited arbitrage by speculators, futures prices adjust to

long or short excess hedging demand. However, the substantial presence of speculators in China suggests that risk absorption capacity is unlikely to be an issue. On the other hand, the currency

premium is significant when futures returns are denominated in US dollars. The inflation premium is positive in a sample predominately made up of foodstuffs—a major component of the CPI basket

in China.

7

Those small‐scale commodities traded during 1994–1998 experienced over‐speculation by large speculators, which drove up the market risk (Xin, Chen, & Firth, 2006). Soybean meal, one of the early

contracts, was set as 10 tons per contract, while the recently launched thermal coal was designed as a big‐scale contract—100 tons per contract.

FAN AND ZHANG

|

673