The impact of oil prices on trade

• Author: Simeon Nanovsky
• Published date: 01 February 2019
• DOI: http://doi.org/10.1111/roie.12383
• Date: 01 February 2019

Rev Int Econ. 2019;27:431–447. wileyonlinelibrary.com/journal/roie © 2018 John Wiley & Sons Ltd

|

431

1

|

INTRODUCTION

In macroeconomics oil price changes, being motivated by the stagflation period of the 1970s, have

generally been associated with the role they play in the overall economy. The literature treats increases

in the oil price as contemporaneously exogenous shocks and then uses VARs to study the economy’s

response to these shocks.1 Unlike this mainstream literature, the aim of this study is to explore the way

oil prices alter the patterns of global trade. This research considers oil prices from the perspective of

the trade literature where they can be seen as influencing trade costs.2 In theory, changes in trade costs

will only alter the allocation or the distribution of trade and therefore the way oil prices influence trade

through their short run impact on the economy is beyond the scope of this paper.3 Nonetheless, any

changes through the economy should not really impact the trade distribution results presented here.

In present times, crude oil is the most significant driving force behind world transportation, supply-

ing 93% of its energy needs (IEA, 2012). Since the 2000s, world oil prices have dramatically increased

Received: 29 January 2018

|

Revised: 5 October 2018

|

Accepted: 8 October 2018

DOI: 10.1111/roie.12383

ORIGINAL ARTICLE

The impact of oil prices on trade

Simeon Nanovsky

Graduate School of Public

Policy,Nazarbayev University, Astana,

Republic of Kazakhstan

Correspondence

Simeon Nanovsky, Graduate School of Public

Policy, Nazarbayev University, 53 Kabanbay

Batyr Ave., Block C3, Astana 010000,

Republic of Kazakhstan.

Email: simeon.nanovsky@nu.edu.kz

Abstract

This paper introduces an oil price–distance interaction

variable in a gravity equation to explain how global trade

behaves as a result of oil price changes. The findings are

that as oil prices increase, international trade becomes

more localized in that countries begin trading relatively

more with their neighbors. In contrast, when they de-

crease, trade becomes more dispersed in that the distance

between countries becomes less relevant. These results are

highly significant across specifications, and the magnitude

is not to be ignored. According to the full specification an

oil price halving will make trade more dispersed by rela-

tively increasing trade by 40% for a distance of 10,000

miles and by 25% for a distance of 1,000 miles.

JEL CLASSIFICATION

F1, F3, F4

432

|

NANOVSKY

from an average of U.S.$20.00 a barrel in 1999 (in 2005 dollars) to reaching over U.S.$90.00 a barrel

in 2008, and since 2015 they have significantly fallen (see Figure 1). Despite these drastic changes,

surprisingly little to no notable research has formally investigated crude oil’s impact on global trade.4

It is reasonable to assume that oil prices affect trade costs, in particular shipping, and thus changes

in shipping will likely drive changes in trade. In a regression, Hummels (2007) confirms that the

fuel costs play an important role in the determination of ad valorem freight costs. Then, the literature

has shown that shipping costs significantly influence trade (Geraci & Prewo, 1977; Harrigan, 1993;

Hummels, 1999; Hummels, Lugovskyy, & Skiba, 2009). Further, if shipping costs are assumed to be

(partly) proxied by distance, it is well known that they play a central role in trade behavior.

Trade might be impacted in different ways. The gravity model of trade can be used for theoretical

predictions.5 One possibility is that a change in the oil price leads to the same percentage change in

shipping/trade costs for all countries. However, if this is the case then theoretically there should be no

change in trade.6 Alternatively, oil prices might have a different impact on trade costs (and therefore

trade) depending on distance. This paper argues that as oil prices rise, countries will tend to increase

trade with their neighbors relative to countries further away. This way trade becomes more localized

relative to before. In contrast, when oil prices fall, distance becomes less of a factor and trade becomes

more dispersed. This is a reasonable hypothesis; as shipping costs are generally more significant por-

tion of a product’s price for longer distances, changes in shipping will likely have a greater effect on

the product’s final delivered price for the longer distance. The gravity model of trade is then able to

explain how these changes in shipping produce inverse changes in trade depending on the distance.

This distributional oil price effect is formally tested here by introducing an oil price–distance

interaction variable in a panel fixed effects gravity regression using all world countries. This type of

regression controls for all time invariant country‐pair specific unobservables, and therefore, exploits

the within country‐pair variation of trade over time. Further, year dummies are also included to control

for overall‐year specific world trade effects. The interaction variable then does not capture absolute

trade changes but just relative trade changes owing to the different country‐pair distances. In other

words, the term captures the magnitude of a change in world trade dispersion as a result of a change

in the oil price. In all the specifications, the interaction effect turns out to be significant and with the

expected sign, implying that global trade becomes more localized when oil prices increase.

A recent paper, Von Below and Vézina (2016) also studies the effect of oil prices on trade. This

research is a natural continuation of their paper. The major additional contributions made here are:

FIGURE 1 Average annual oil price

Ϭ

ϮϬ

ϰϬ

ϲϬ

ϴϬ

ϭϬϬ

ϭϮϬ

ϭϵϱϮ

ϭϵϱϱ

ϭϵϱϴ

ϭϵϲϭ

ϭϵϲϰ

ϭϵϲϳ

ϭϵϳϬ

ϭϵϳϯ

ϭϵϳϲ

ϭϵϳϵ

ϭϵϴϮ

ϭϵϴϱ

ϭϵϴϴ

ϭϵϵϭ

ϭϵϵϰ

ϭϵϵϳ

ϮϬϬϬ

ϮϬϬϯ

ϮϬϬϲ

ϮϬϬϵ

ϮϬϭϮ

ZĞĂů;ϮϬϬϱͿŽůůĂƌƐWĞƌĂƌƌĞů

ƵďĂ

ŝ

td/

ƌĞŶƚ