The cross‐predictability of industry returns in international financial markets

• Published date: 01 December 2023
• Author: Xin Wang,Haofei Zhang
• Date: 01 December 2023
• DOI: http://doi.org/10.1111/irfi.12426

ORIGINAL ARTICLE

The cross-predictability of industry returns

in international financial markets

Xin Wang

1

| Haofei Zhang

2

1

School of Economics, Shandong University,

Jinan, China

2

School of Finance, Nankai University,

Tianjin, China

Correspondence

Xin Wang, School of Economics, Shandong

University, 27 Shanda Nanlu, Jinan 250100,

China.

Email: xin.wang@sdu.edu.cn

Abstract

This article finds evidence of return cross-predictability

among trading partners in international financial markets.

We show that the predictability of international customers

dominates the predictability of domestic customers, and the

predictability of international intra-industry customers dom-

inates the predictability of international inter-industry cus-

tomers. This return cross-predictability decreases with two

country characteristics: financial sophistication and size.

KEYWORDS

inter-industry trading partners, international financial markets,

intra-industry trading partners, return predictability

1|INTRODUCTION

In the post-World War II era, international trade has grown tremendously; the sum of exports and importsmeasured

as a share of gross domestic product (GDP) grew from 25.0% in 1960 to 57.7% in 2015. (The World Bank, 2016).

Given the increasing importance of international trade, it is critical to evaluate how value-relevant information is dif-

fused in trade-linked stock markets.

In this article, we examine how quickly an industry's stock returns respond to information about its trading

partners across countries and industries. In efficient markets, all customer information is immediately incorpo-

rated into the returns of the corresponding suppliers, without any lead–lag relationship between the customers

and suppliers' returns. However, we find evidence that market frictions delay the incorporation of customer

information into corresponding supplier returns. The lagged monthly returns of international customers can pre-

dict the contemporaneous returns of corresponding suppliers at the country-industry level, suggesting that mar-

ket frictions delay the diffusion of material information in international financial markets. The cross-predictability

of lagged customer returns remains robust when we control for lagged long-run customer returns and lagged

short-run and long-run supplier returns. We also find that international customer and supplier returns cross-

predict each other.

1

Received: 22 November 2021 Revised: 18 April 2023 Accepted: 7 July 2023

DOI: 10.1111/irfi.12426

© 2023 International Review of Finance Ltd.

International Review of Finance. 2023;23:859–885. wileyonlinelibrary.com/journal/irfi 859

When comparing return predictability at the intra-industry and inter-industry levels, we find that predictability

at the intra-industry level is much more important than predictability at the inter-industry level. For a particular sup-

plier, we classify all corresponding customers into three groups: customers in the same industry but a different coun-

try from the supplier (international intra-industry customers), customers in the same country but a different industry

from the supplier (domestic inter-industry customers), and all other customers from different industries and different

countries (international inter-industry customers). We find that return predictability is the largest, in terms of both

statistical and economical significance, for international intra-industry customers. This result suggests that additional

information from international intra-industry trading partners is more valuable than information from international

inter-industry and domestic inter-industry trading partners.

We also divide customers into foreign and domestic categories. We find that the predictability of foreign cus-

tomers dominates, suggesting the more gradual diffusion of foreign information than domestic information. There

may not be additional valuable information from domestic inter-industry customers in the same month, as most of

the domestic information is incorporated into stock prices at monthly, or even shorter, frequencies. Accordingly,

international information is more important. We further classify international customers by industry and find that

intra-industry customers create more valuable information than inter-industry customers. Return predictability

remains when we pool all of the customers together. Furthermore, to emphasize the economic link between sup-

pliers and customers, we randomly assign customers (and weights) from the industry-country observations that are

not the actual customers of the focal supplier. We find that return predictability disappears when we shuffle trading

partners and use trading weights based on randomly matched suppliers and customers; this finding indicates that our

results are not spurious.

We create a long-short trading strategy based on this return cross-predictability of customers for their

corresponding suppliers. The profits from this trading strategy are both statistically and economically significant. A

value-weighted portfolio of corresponding suppliers yields monthly abnormal returns of 0.539%.

This return cross-predictability decreases with the following country characteristics of suppliers: financial sophis-

tication and size. We use the depth of the financial markets, which is defined as market capitalization divided by

GDP, as a proxy for financial sophistication in exporting countries. For suppliers in countries with higher levels of

financial sophistication, information about customers is more quickly incorporated into the corresponding supplier

stock prices than in countries with lower levels of financial sophistication, thereby generating less return cross-

predictability of those customers. Furthermore, size, as measured by market capitalization in exporting countries,

plays a vital role in explaining the observed customer return cross-predictability.

We extend the study of Menzly and Ozbas (2010) to an international scale and use a sample of OECD input–

output (ICIO) tables from 52 countries in our empirical analysis. Our article is primarily related to the substantial liter-

ature on return predictability due to gradual information diffusion across geographic segments, firms, and time.

Cohen and Frazzini (2008) find that the returns of customer firms predict the returns of supplier firms. Menzly and

Ozbas (2010) show that stocks in supplier and customer industries cross-predict each other's returns in the

United States. Cohen and Lou (2012) find that standalone pure players predict the stock returns of conglomerates.

Nguyen (2012) and Huang (2015) show that value-relevant foreign information slowly diffuses into the stock prices

of US multinational firms. Rizova (2013) shows that the market returns of a country's major trading partners predict

that country's stock market returns. Cao et al. (2016) document that the stock returns of strategic alliance partners

predict each other. Cen et al. (2016) focus on the implications of the institutional and regulatory environment of for-

eign customers for US suppliers. Finke and Weigert (2017) find that foreign information only gradually dilutes into

the stock prices of multinational firms worldwide. Smajlbegovic (2019) finds that information about regional eco-

nomic activity is gradually incorporated into stock prices. Using global cross-firm ownership data, Chang et al. (2022)

find predictability in complex ownership firms. Our article complements the literature by finding return predictability

among international intra-industry and inter-industry trading partners.

Our article is also closely related to the literature that emphasizes the importance of industry affiliation for

return predictability (Asness and Stevens (1995); Moskowitz and Grinblatt (1999); Hong et al. (2007); Hou (2007);

860 WANG and ZHANG