The predictability of iron ore futures prices: A product‐material lead–lag effect

• Published date: 01 September 2023
• Author: Mengxi He,Yudong Wang,Yaojie Zhang
• Date: 01 September 2023
• DOI: http://doi.org/10.1002/fut.22440

Received: 26 January 2023

|

Accepted: 28 May 2023

DOI: 10.1002/fut.22440

RESEARCH ARTICLE

The predictability of iron ore futures prices:

A product‐material lead–lag effect

Mengxi He |Yudong Wang |Yaojie Zhang

School of Economics and Management,

Nanjing University of Science and

Technology, Nanjing, China

Correspondence

Yaojie Zhang, School of Economics and

Management, Nanjing University of

Science and Technology, Xiaolingwei 200,

210094 Nanjing, Xuanwu District, China.

Email: yaojie_zhang@126.com

Funding information

National Natural Science Foundation of

China, Grant/Award Numbers: 72001110,

71722015, 72071114

Abstract

This study investigates the lead–lag effects between product futures and

raw material futures. Results show that returns on product futures lead

returns on raw material futures: lagged product futures returns can

significantly predict raw material futures returns in‐and out‐of‐sample.

This product‐material lead–lag effect is mainly driven by bad news and is a

short‐term phenomenon. Moreover, returns on product futures, especially

those based on bad news, can provide substantial economic gains to

investments in raw material futures. The lead–lag effect is associated with

frictions in the information diffusion from product futures markets to raw

material futures markets and the slower response of raw material futures

markets to common market information.

KEYWORDS

futures market, information frictions, lead–lag effects, limited attention, return predictability

JEL CLASSIFICATION

C22, C53, C58, G17

1|INTRODUCTION

Prices of assets with common economic fundamentals tend to fluctuate together. However, empirical results show

that the price movements of these interconnected assets are not perfectly synchronized. This phenomenon is

known as the lead–lag effect and many studies have provided evidence (see, e.g., Hong et al., 2007;Lo&

MacKinlay, 1990; Parsons et al., 2020). Similarly, we explore a new type of lead–lag relationship in this study,

namely, the product‐material lead–lag effect. The results show that the futures prices on the product‐side lead that

on the raw material side.

We first investigate the lead–lag pattern between product futures and raw material futures. On the one hand,

analysts and investors are more focused on segmented markets. As a result, information reacts more quickly in more

segmented markets, resulting in segments leading the overall market (see, e.g., Hong et al., 2007; Jiang et al., 2019). In

our case, the product futures market is a more segmented market relative to the raw materials futures market.

1

On the

other hand, the consumer side of the market can lead the supply side of the market (see, e.g., Cohen & Frazzini, 2008;

Li et al., 2021). In our case, product futures are on the consumer side. In summary, the existing literature implies that

product futures lead raw material futures.

J Futures Markets. 2023;43:1289–1304. wileyonlinelibrary.com/journal/fut © 2023 Wiley Periodicals LLC.

|

1289

1

In our case, iron ore futures correspond to multiple product futures.

The common explanation for these lead–lag effects is that investors' limited attention and limited

information‐processing capabilities lead to the existence of information frictions (see, e.g., Hong et al., 2007;

Hou, 2007; Rapach et al., 2013). In this study, we consider two types of information propagation frictions to

explain the product‐material lead–lag effect. The first type of information comes from the leader, product futures

in our case. In this setup, there is friction in the diffusion of information from the product side to the raw

material side. Therefore, the prices of raw material futures and product futures are not synchronized, but rather

raw material futures lag behind product futures. This fact generates the lead–lag effect between the product and

raw material. The second type of information comes from the common market for the asset under consideration.

In this setup, product futures and raw material futures react differently to the common information. That is, the

lagger (in our case, raw material futures) price adjusts more slowly to this common information and thus the

product futures market leads the raw material futures market. Corresponding empirical results provide evidence

for the existence of both types of information.

When lagged product futures returns are used as predictors, we find that product‐side returns can positively

and significantly predict future raw material‐side returns. And, this forecasting power is mainly driven by bad

news and is a short‐term phenomenon. In addition, we find that product futures returns can provide useful

information for out‐of‐sample forecasting of raw material futures returns. Finally, the results show that

investors who forecast raw material futures returns based on product futures returns can achieve substantial

economic returns.

The remainder of the study is organized as follows. Section 2reviews the relevant literature and highlights our

contribution. Section 3describes our data. Section 4conducts the empirical analyses. Section 5provides out‐of‐sample

evidence. Section 6provides potential explanations. Finally, Section 7concludes.

2|RELEVANT LITERATURE AND OUR CONTRIBUTION

In this section, we review the relevant literature on the lead–lag effect in financial markets and the predictability of

futures prices. In addition, we separately discuss the contribution of this study in the two aspects.

2.1 |Lead–lag effect in financial markets

Researchers conduct numerous studies and provide empirical evidence on the lead–lag effect in financial

markets. Atchison et al. (1987) earlier study serial correlation in portfolio returns. Lo and MacKinlay (1990)find

that returns of small firms are correlated with past returns of large firms, demonstrating the existence of the

lead–lag effect between firms of different sizes. Later, Hou (2007) finds that the lead–lag effect between large and

small firms is primarily an intraindustry phenomenon. Hong et al. (2007) demonstrate that many industries,

such as retail, services, commercial real estate, metals, and oil, lead the overall stock market. Returns in these

industries significantly predict future returns in the overall stock market. Cohen and Frazzini (2008)findthe

lead–lag relationship between the stock returns of companies in the supply chain and Cohen and Lou (2012)

document the underreaction between focused firms and conglomerates. Jacobsen et al. (2019) find that the

industrial metals markets lead the stock market, although the effect varies across economic periods. Ali and

Hirshleifer (2020)examinethelead–lag effect for companies with common analysts and Parsons et al. (2020)

document the lead–lag effect on returns between coheadquartered firms in different sectors. Huang et al. (2022)

find that the customer‐supplier lead–lag effect exists when information from customer firms is sufficiently

continuous. Complementary to their study, we explore lead–lag effects between product futures and raw

material futures and find that returns on product futures lead returns on raw material futures.

2.2 |Predictability of futures prices

The prediction of commodity futures prices has received extensive attention from scholars in recent years.

Ewees et al. (2020) integrate chaotic behavior into a recent meta‐heuristic method grasshopper optimization

algorithm and forecast iron ore price. Zhang et al. (2021) use a variety of machine learning models and find that

1290

|

HE ET AL.