Can skewness of the futures‐spot basis predict currency spot returns?

• Date: 01 November 2019
• Author: Liyan Han,Libo Yin,Xue Jiang
• DOI: http://doi.org/10.1002/fut.21991
• Published date: 01 November 2019

Received: 30 November 2018

|

Revised: 9 December 2018

|

Accepted: 9 December 2018

DOI: 10.1002/fut.21991

RESEARCH ARTICLE

Can skewness of the futures‐spot basis predict currency

spot returns?

Xue Jiang

1

*

|

Liyan Han

1

*

|

Libo Yin

2

*

1

School of Economics and Management,

Beihang University, Beijing, China

2

School of Finance, Central University of

Finance and Economics, Beijing, China

Correspondence

Libo Yin, School of Finance, Central

University of Finance and Economics,

39 South College Road, Haidian District,

100081 Beijing, China.

Email: yinlibowsxbb@126.com;

0020130053@cufe.edu.cn

Funding information

Program for Innovation Research in

Central University of Finance and

Economics; National Natural Science

Foundation of China, Grant/Award

Numbers: 71671193, 71871234, 71371022

Abstract

This paper examines the relationship between skewness of the futures‐spot

basis and expected currency spot returns. The empirical results show that the

expected spot returns are negatively correlated with the basis skewness. We find

that the basis skewness exhibits statistically significant in‐sample and out‐of‐

sample forecasting power. Furthermore, the basis skewness beat the random

walk (without drift) in economic measures. The impacts of the basis skewness

on spot returns barely vary with time and have no structural breaks. We also

find that the basis skewness can really improve the predictability of spot

returns, even when the futures‐spot basis is considered.

KEYWORDS

currency spot returns, futures‐spot basis, out‐of‐sample forecasts, skewness, time‐varying risk

premium

1

|

INTRODUCTION

As early as Rubinstein (1973) and Kraus and Litzenberger (1976, 1983), theoretical studies have developed models of

expected return that incorporate skewness. As indicated by Harvey and Siddique (2000), Mitton and Vorkink (2007), and

Barberis and Huang (2008), if asset returns have a systematic skewness risk, expected returns should include rewards for

bearing this type of risk; they also indicate a negative and significant relationship between expected returns and

conditional skewness. However, empirical results are mixed. Some previous studies (Amaya, Christoffersen, Jacobs, &

Vasquez, 2015; Bali & Murray, 2013; Bali, Cakici, & Whitelaw, 2011; Chang, Christoffersen, & Jacobs, 2013; Conrad,

Dittmar, & Ghysels, 2013; Eraker & Ready, 2015) find evidence in support of this negative relationship, whereas

Cremers and Weinbaum (2010), and Xing, Zhang, and Zhao (2010) document a positive relationship. Given these mixed

results, a study that focuses on a different asset class can shed light on the relationship between skewness and expected

returns.

In this paper, we try to shed new light on these mixed results by focusing on foreign exchange or currency markets

(FX). We choose this market for a number of reasons. First and foremost, we are inspired by the unique market features

and the types of investors in FX markets. On the one hand, there are no short‐selling restrictions in FX markets.

According to the theory on skewness preferences proposed by Mitton and Vorkink (2007), skewness matters because of

investors’preferences for positive skewness (lottery‐type payoffs), which causes positively skewed assets to become

overpriced and earn lower expected returns than assets with negative skews. This overpricing is not arbitraged away

because of short‐selling restrictions, which do not exist in FX markets. On the other hand, since currency markets are

dominated by speculators and hedgers, with retail investors rarely participating, the role of skewness is more in line

J Futures Markets. 2019;39:1435–1449. wileyonlinelibrary.com/journal/fut © 2019 Wiley Periodicals, Inc.

|

1435

*Xue Jiang, Liyan Han, and Libo Yin have contributed equally to this paper.