Asymmetric effects of oil price changes on the balance of trade: Evidence from selected African countries

• Published date: 01 November 2019
• Author: Jungho Baek,Kyoung Doug Kwon
• Date: 01 November 2019
• DOI: http://doi.org/10.1111/twec.12844

World Econ. 2019;42:3235–3252. wileyonlinelibrary.com/journal/twec

|

3235

© 2019 John Wiley & Sons Ltd

Received: 12 November 2018

|

Revised: 31 May 2019

|

Accepted: 25 June 2019

DOI: 10.1111/twec.12844

ORIGINAL ARTICLE

Asymmetric effects of oil price changes on the

balance of trade: Evidence from selected African

countries

JunghoBaek1

|

Kyoung DougKwon2

1Economics,University of Alaska Fairbanks, Fairbanks, Alaska

2Korea Development Institute, Sejong, Korea

KEYWORDS

Africa, asymmetry approach, crude oil, trade balance

1

|

INTRODUCTION

Many scholars have actively explored the influences of crude oil price changes on a country's economic

activity over the past three decades. Given the scope of the research, these studies can be roughly

split into two types of categories. The first category contains numerous articles that consider how the

price of crude oil has an effect on a country's domestic macroeconomic factors. Some, for example,

probe the influences of oil price changes on economic growth (Burbidge & Harrison, 1984; Ferderer,

1996; Gisser & Goodwin, 1986; Herrera, Lagalo, & Wada, 2015; Huang, Hwang, & Peng, 2005;

Jimenez‐Rodriguez & Sanchez, 2005; Jones, Leiby, & Paik, 2004; Killian & Vigfusson, 2011; Lardic

& Mignon, 2006) and employment (Altay, Topcu, & Erdogan, 2013; Michieka & Gearhat, 2019; Uri

& Boyd, 1996). Others address inflation, interest rates and stock market issues arising from fluctua-

tions in crude oil prices (Albulescu, Oros, & Tiwari, 2017; Basher, Haug, & Sadorsky, 2012; Boyer &

Filion, 2007; Elyasiani, Mansur, & Odusami, 2011; Faff & Brailsford, 1999; Huang, Masulis, & Stoll,

1996; Jones & Kaul, 1996; Nazarian & Amiri, 2014; Sadorsky, 1999; Salisu & Isah, 2017; Valcarcel &

Wohar, 2013; Zhu, Li, & Li, 2014). In general, they employ country‐specific data and time series mod-

els and uncover that domestic macroeconomic variables seem to be susceptible to oil price changes.

Since the seminal work done by Backusab and Crucinic (2000), the second body of the research

has lately been growing that analyses the impact of oil shocks on a country's trade balance. Examples

contain Backusab and Crucinic (2000), Kilian, Rubucci, and Spatafora (2009), Bodenstein, Erceg, and

Guerrieri (2011), Le and Chang (2013), Rafiq, Sgro, and Apergis (2016) and Yalta and Yalta (2017).

Kilian et al. (2009) apply a structural VAR (SVAR) technique to various oil‐exporting countries and

oil‐importing countries; they discover that crude oil price changes seem to be one of the vital factors

3236

|

BAEK And KWOn

influencing external balances. Rafiq et al. (2016) recently expand the work of Kilian et al. (2009) by

employing a non‐linear panel method to investigate an asymmetric response to oil price changes; they

reveal that oil price plunges have a positive impact on oil‐exporting countries, whereas a steady oil

price has a more favourable impact on oil‐importing countries than a price decline. A crucial weak-

ness of most empirical studies in the second category is that fluctuations in oil prices are assumed to

have symmetric impacts. This indicates that if a 1% upsurge in the price of oil triggers the balance

of trade to increase by, say, x%, a 1% decline in crude oil prices must worsen it by x%. Otherwise, oil

price changes are said to have asymmetric impacts. In fact, since expectations and responses of ex-

porters and importers to an oil price upsurge are highly likely to be different from those to an oil price

decline, oil price changes are more likely to affect external balances asymmetrically in the real world

as Rafiq et al. (2016) argue and demonstrate. Studies to date, however, do not consider an asymmetric

response to oil price changes in their empirical analyses and raise questions about the validity of their

conclusions. Further, attention in the second category has typically been on the aggregated trade bal-

ances such as eight OECD countries (Backusab & Crucinic, 2000) and three Asian economies (Le &

Chang, 2013) when assessing the oil price–trade balance nexus. However, one country's trade surplus

with some trade partners could be dominated by trade deficits with others (and vice versa) in such

a model. Even countries with seemingly similar economies are oftentimes likely to exhibit different

trade patterns with various trading partners. Accordingly, the empirical results of these studies may

suffer from aggregation bias of data, thereby providing misleading results. Hence, there remains a

strong need for improved analysis of the oil price impacts on the balance of trade.

The contribution of this paper is to extend the second category of the literature by considering the

asymmetric effects of changes in crude oil prices on the balance of trade in the framework of individ-

ual countries, particularly six major economies in Africa—Nigeria, South African, Egypt, Algeria,

Morocco and Kenya (Table 1). For this, we utilise a non‐linear autoregressive distributed lag (NARDL)

method, being originally proposed by Shin, Yu, and Greenwood‐Nimmo (2014).1 It is worth pointing

out that numerous studies have attempted to isolate the independent impact of oil prices on economic

1 It is worth mentioning that the University of Wisconsin‐Milwaukee economist Mohsen Bahmani‐Oskooee is the leading

expert in the implementation of the NARDL approach. For example, Bahmani‐Oskooee et al. (2017) use the NARDL to study

the short‐ and long‐run asymmetric impacts of exchange rate changes on the UK trade balance with its 11 trading partners.

TABLE 1 Top 10 African economies (2015)

Country GDP (billion $)

Exports (million $) Imports (million $)

Oil Non‐oil Oil Non‐oil

Nigeria 594.3 84,172.5 10,046.4 11,947.4 65,523.5

South Africa 341.2 32,510.9 96,650.4 29,861.2 134,187.1

Egypt 284.9 8,829.7 30,962.1 25,582.1 116,762.7

Algeria 227.8 66,544.3 2,636.6 5,473.5 112,745.6

Morocco 112.6 2,706.2 44,812.4 12,649.1 66,194.7

Sudan 115.6 1,259.3 3,575.2 475.2 18,213.2

Kenya 53.4 1,444.7 8,967.2 5,106.0 29,547.5

Angola 49.9 64,885.7 104.3 595.6 28,747.7

Libya 49.3 19,746.0 482.5 485.9 27,847.3

Tunisia 49.1 2,226.2 26,183.6 6,459.1 36,776.9

Note: Data sources: IMF and WTO.