Measuring the Real and Financial Connectedness of Selected African Economies with the Global Economy

• Author: Gladys C. Aneke,Jonathan E. Ogbuabor,Oyun Erdene‐Urnukh,Anthony Orji
• DOI: http://doi.org/10.1111/saje.12135
• Date: 01 September 2016
• Published date: 01 September 2016

MEASURING THE REAL AND FINANCIAL

CONNECTEDNESS OF SELECTED AFRICAN ECONOMIES

WITH THE GLOBAL ECONOMY

JONATHAN E.OGBUABOR*, ANTHONY ORJI,GLADYS C.ANEKE AND OYUN ERDENE-URNUKH

Abstract

We examine the real and financial connectedness of selected African economies with the global

economy using a network approach. We find that the connectedness of African economies with

the global economy is quite sizable, with the global financial crisis increasing the connectedness

measures above their pre-crisis levels. The results show that U.S., EU and Canada dominate Afri-

ca’s equity markets, while China, India and Japan dominate Africa’s real activities. Our results

suggest that African economies are predominantly small open economies, deeply interconnected

but systemically unimportant and vulnerable to headwinds emanating from the dominant econo-

mies in the overall global economy.

JEL Classification: F02, C53, C32, F36, N27

Keywords: Connectedness, network approach, vector autoregressive (VAR) model, global financial

crisis, African economies

1. INTRODUCTION

Several studies have examined the international macro-economic linkages or commonal-

ity in macro-economic fluctuations of entities in the global economy (examples include

Kose et al., 2003; Doyle and Faust, 2005; Canova et al., 2007; Diebold and Yilmaz,

2009, 2014, 2015a; Park and Shin, 2014; Greenwood-Nimmo et al., 2015, henceforth

GNS25). These studies became necessary following the rising international linkages

among entities in the globe in recent decades, and the need to better model and under-

stand the dynamics of these connections. Much of the rise in global real and financial

interconnectedness has been attributed to a number of factors, mainly the current wave

of globalisation sweeping across the world; the rising level of deregulation, privatisation

and liberalisation across the globe; the increasing level of specialisation in production

among countries owing mainly to differences in technologies and endowments; and

breakthroughs in information and communications technology (Park and Shin, 2014).

Park and Shin (2014) stressed that though interconnectedness among economies

could engender shared prosperity, it nonetheless encourages the rapid propagation of real

and financial contagion through the global economic and financial system. A case in

point is the recent 2007–2008 Global Financial Crisis (GFC), which led to decline in

output growth and stock market crashes especially among African economies. There are

two important lessons from this crisis. One, it has indeed shown that macro-economic

linkages between African economies and the rest of the world are important and deserve

* Corresponding author: Lecturer, Department of Economics, University of Nigeria, Nsukka,

Enugu State 410108, Nigeria. E-mail: jonathan.ogbuabor@unn.edu.ng

V

C2016 Economic Society of South Africa. doi: 10.1111/saje.12135

364

South African Journal of Economics Vol. 84:3 September 2016

South African Journal

of Economics

to be studied. Two, these macro-economic linkages arise mainly through real connected-

ness that highlights the cross-country business cycle comovement and financial connect-

edness that shows how financial contagion is spread across the globe (GNS25; Dees

et al., 2007). However, apart from real and financial connectedness, Diebold and Yilmaz

(2014) document other forms of linkages among entities in the global economy.

Majority of the studies evaluating the macro-economic connectedness among entities

in the global economy have thus far concentrated on the highly industrialised economies

like the U.S., UK, Eurozone, China, Japan, among others. For instance, Diebold and

Yilmaz (2015a) focused on the highly advanced G-6 countries (i.e. G-7 less Canada);

Diebold and Yilmaz (2014) focused on the U.S. financial institutions; Doyle and Faust

(2005) and Stock and Watson (2005) both focused on the G-7 economies; Yilmaz

(2009) focused on the East Asian stock markets; while GNS25 evaluated 25 countries/

regions including South Africa. Clearly, the existing literature appears quite scanty on the

macro-economic connectedness of African economies with the rest of the world. It is the

goal of this study to fill this gap in the literature.

In view of the foregoing, this paper evaluates the real and financial connectedness of

selected African economies with the global economy over the period 1981Q1–2012Q3,

based on three specific objectives. First, the paper measures and analyzes the real and finan-

cial connectedness of African economies with the global economy. Second, the paper deter-

mines which of the countries in the global economy exert the most dominant influence on

African economies. Third, the paper examines the impact of the 2007–2008 GFC on the

real and financial connectedness of African economies with the global economy. Following

established literature (e.g. Park and Shin, 2014; Diebold and Yilmaz, 2015a;GNS25), it is

expected that African economies should be categorised as small open economies that are vul-

nerable to shocks emanating from the highly industrialised economies.

To achieve the above objectives, we adopt the recent connectedness-measurement

technology developed by Diebold and Yilmaz (2009), which is rooted in network

theory.

1

The normalised generalised forecast error variance decompositions (NGFEVDs)

from the underlying vector autoregressive (VAR) model is used to build both directional

and non-directional connectedness measures at various levels of aggregation. Pesaran and

Shin (1998) note that the NGFEVDs obtained directly from the estimated parameters

and covariance matrix of the VAR model are invariant to variable ordering and do not

require any additional restrictions beyond those needed for estimation and identification.

Contrary to Diebold and Yilmaz (2015a) and GNS25, our results indicate that the

roles of India and UK in the global economy can no longer be called unimportant when

African economies are the focus of study. In addition, the results reveal that the connect-

edness of African economies with the global economy is quite substantial and that the

emerging economic structure after the GFC indicates a dwindling role for the U.S. in

Africa and an increasing role for Asian giants including India, China, and Japan. Among

other interesting findings, the results further show that the role of the domestic economy

in African equity markets declined after the GFC, suggesting loss of confidence by

Africans in their own equity markets. Overall, the results are consistent with the

1

GNS25 generalised and extended this approach to the multi-country multi-variable case. As a

pioneer study providing a benchmark analysis and accounting substantially for the African econo-

mies, this paper adopts the univariate approach of Diebold and Yilmaz (2009). Subsequent analy-

sis will be extended to the multi-variate case of GNS25 underpinned by the global VARmodel.

365South African Journal of Economics Vol. 84:3 September 2016

V

C2016 Economic Society of South Africa.

characterisation of African economies as small open economies that are predominantly

shock receivers and systemically unimportant in the global economic arena.

The remainder of this paper proceeds as follows. Section 2 presents an overview of the

empirical literature, based on the network approach. The research methodology is pre-

sented in Section 3, detailing the model specification, the GFEVDs and the construction

of the generalised connectedness measures (GCMs). Section 4 presents the data and

some preliminary descriptive data analysis. The empirical results of this paper are dis-

cussed in Section 5, while Section 6 concludes the paper.

2. OVERVIEW OF THE EMPIRICAL LITERATURE

The review of empirical literature here considers mainly the studies that are based on the

network approach developed by Diebold and Yilmaz (2009). However, other studies on

the macro-economic linkages among entities in the global economy are also highlighted,

especially the multi-disciplinary ones involving network theory. Studies employing the

Diebold and Yilmaz (2009) methodology share certain common characteristics, namely:

(1) they are generally based on connectedness measures distilled from FEVDs of an

approximating VAR model; (2) they measure the direction and strength of linkages

among entities in the system; (3) they can identify systemically important entities in the

system; (4) they can study the dynamic nature of shock propagation among entities in

complex systems.

In their pioneer paper, Diebold and Yilmaz (2009) analysed 19 global equity markets

over the period 1992M1–2007M11, and find that return spillovers display a gently

increasing trend but no bursts, whereas volatility spillovers display no trend but clear

bursts. Following this initial paper, Diebold and Yilmaz have also applied this methodol-

ogy in a series of other publications. Diebold and Yilmaz (2012) examined four main

classes of U.S. assets markets (stocks, bonds, foreign exchange and commodities markets)

between 1999M1 and 2010M1. They find that volatility fluctuations were significant in

all four markets; however, cross-market volatility transmissions were very limited until

the 2007–2008 GFC began. In addition, volatility spillovers increased with the intensity

of the GFC, and the direction of transmission was mainly from the stock market to other

markets, especially following the collapse of Lehman Brothers in September 2008.

Diebold and Yilmaz (2014) measured the connectedness of U.S. financial firms using

equity returns volatilities data between 1999M5 and 2010M 4. The findings reveal that

financial linkages are important in the transmission of shocks among economic entities;

and high total connectedness index of 78%, which is higher than the connectedness they

found previously among international stocks. Diebold and Yilmaz (2015a) measured the

real connectedness of six advanced economies (i.e. G-7 less Canada) using industrial pro-

duction data for 1958M1–2011M12. The findings indicate that global output connect-

edness is sizable and time-varying over the business cycle, and that the U.S. and Japan

are the main transmitters of output shocks.

Apart from Diebold and Yilmaz, other researchers have also applied the network

methodology. Park and Shin (2014) evaluated the connectedness of the Korean economy

with the global economy between 1980Q2 and 2012Q3 using an underlying Global

VAR model. The findings indicate that the U.S., Europe, China, the ASEAN group and

the global energy markets exert dominant influence on the Korean economy. Domestic

conditions were found to be important on the short to medium term, whereas external

366 South African Journal of Economics Vol. 84:3 September 2016

V

C2016 Economic Society of South Africa.