Exporting to fragile states in Africa: Firm‐level evidence

• Author: Peter Wankuru Chacha,Lawrence Edwards
• DOI: http://doi.org/10.1111/rode.12408
• Published date: 01 August 2019
• Date: 01 August 2019

SPECIAL ISSUE ARTICLE

Exporting to fragile states in Africa: Firm-level

evidence

Peter Wankuru Chacha

1

|

Lawrence Edwards

2

1

The World Bank Group-Kenya,

NAIROBI, Kenya

2

School of Economics, University of

Cape Town, Rondebosch, South Africa

Correspondence

Peter Wankuru Chacha, The World Bank

Group-Kenya country office Delta Center

building, 17th floor, room 1783,

NAIROBI, Kenya.

Email: peter.chacha80@gmail.com

Funding information

African Economic Research Consortium

(AERC), Grant/Award Number: RC

16502

Abstract

This study analyses the effect of fragility in destination

markets on firm export behavior and the role of firm size

in mediating adverse outcomes. The analysis is conducted

using firm transaction data on Kenyan exports to Africa

over the period 2004–2013. The analysis reveals that fra-

gility negatively affects a firm’s decision to enter a given

destination market, reducing Kenya’s bilateral trade flows

to African countries. Larger firms are more resilient to

destination shocks in fragility and are less likely to exit.

These results are robust to alternative measures of desti-

nation fragility, and the exclusion of bordering countries

and the East African Community partner states. Our anal-

ysis reveals that the effect of business fragility (regulatory

quality, government effectiveness, and control of corrup-

tion) dominates that of political fragility (voice and

accountability, rule of law, and political stability),

although both effects are negative and significant.

1

|

INTRODUCTION

In this paper, we ask if fragility in a destination market reduces the likelihood of a Kenyan firm

exporting to that destination. For the exporting firm, a change in destination country fragility can

be viewed as a set of exogenous shocks to entering and remaining in that market. This is particu-

larly relevant in the case of firm exports to sub-Saharan African (SSA) countries, many of which

are characterized by high degrees of fragility. Understanding the effect of destination fragility on a

firm’s export decisions and the mechanisms through which firms mediate this effect is thus impor-

tant for trade within SSA countries.

Destination country fragility is a multidimensional concept that is hard to define and mea-

sure precisely.

1

There is no globally accepted definition of fragility, and nations do not like

this label (Chauvet & Collier, 2008). Chauvet and Collier (2008) define a fragile country as a

DOI: 10.1111/rode.12408

Rev Dev Econ. 2019;23:1177–1201. wileyonlinelibrary.com/journal/rode ©2018 John Wiley & Sons Ltd

|

1177

low-income country in which economic policies, institutions and governance are so poor that

growth is highly unlikely. In the World Development Report (WDR) on conflict, insecurity

and development (World Bank, 2011, p. 18), fragile situations are defined as “periods when

states or institutions lack the capacity, accountability or legitimacy to mediate relations

between citizen groups and b etween citizens and the sta te, making them vulnerabl e to

violence.”

Faced with diverse definitions and the fact that fragility covers many different concepts, our

definition of destination fragility follows the WDR definition. This definition is chosen partly

because it was an outcome of a comprehensive review by a committee of experts drawn from sev-

eral disciplines (economics, political science, sociology, etc.) and the process was widely consulta-

tive (World Bank, 2011). It is related but not similar to the definition of economic vulnerability as

captured by the Economic Vulnerability Index (EVI), that is, the likelihood that a country’s devel-

opment could be hindered by unforeseen exogenous shocks (Cariolle, Goujon, & Guillaumont,

2016; Guillaumont, 2009). The extreme variant of fragility definition is also associated with con-

flict and civil wars, a field that has received far more attention in the literature (Qureshi, 2013;

Glick & Taylor, 2010; Martin, Mayer, & Thoenig, 2008; Collier & Hoeffler, 2004; Anderson &

Marcouiller, 2002). This aspect of fragility is not addressed in this paper.

The complexity of defining fragility has resulted in several indices of fragility being used in the

literature. These include the Worldwide Governance Indicators (WGI) (Kaufmann, Kraay, & Mas-

truzzi, 2011), the Country Policy and Institutional Assessment indices (World Bank, 2015), the

International Country Risk Guide (ICRG) index (PRS Group, 2011) and the Fragile States Index

(FSI; Messner, 2005). In this paper, we make use of the six sub-indices of the WGI as these more

closely reflect the definition of fragility proposed by World Bank (2011).

We study in a variety of ways the effect of destination country fragility on Kenyan firms’deci-

sion to export to a given destination in Africa. Firstly, we construct a composite indicator of fragi-

lity using principal component analysis (PCA) and estimate how changes in this indicator

influence the decision by Kenyan firms to export to that country. We then evaluate the effect of

each of the WGI indicators separately, before splitting our composite fragility indicator into two

components, namely, political risk and business risk. This approach allows us to investigate the

channels through which fragility affects firm-level export participation. We also look at whether

firm size attenuates the adverse effect of fragility on the export decision. Finally, drawing on the

decomposition method of Bernard, Redding, and Schott (2006), we estimate how fragility affects

aggregate exports through its influence on the number of exporters, the number of exported prod-

ucts, the average export value per firm, and the number of firm–product observations for which

trade to a given country is positive.

In addressing the above, this paper contributes to the existing literature in three main ways.

Firstly, we exploit the richness of transactions data that enable us to observe firms’destination

choice for their exports to Africa. This allows us to evaluate the effect of destination fragility on

the firms’decision to serve a given destination in Africa with exports as well as the role of firm

size in export markets in mediating this outcome. In doing so, we extend the related international

literature that has largely made use of aggregated trade data that conceal potentially diverse adjust-

ments to destination fragility taking place at the firm and product level (Glick & Taylor, 2010;

Mansfield & Bronson, 1997; Mansfield & Pevehouse, 2000; Penubarti & Ward, 2000; Pollins,

1989; Morrow, Siverson, & Tabares, 1998).

2

Secondly, our study provides an additional explanation for why intra-African trade flows are

low, despite the embracing by governments of regional integration as a core part of trade pol-

icy in Africa (Carrere, 2004; Martin et al., 2008; Yeats, 1998). While the contribution of high

1178

|

CHACHA AND EDWARDS