Does Regional Currency Global Shock Ob Africa PDF

Journal of Economic Studies
Downloaded by COMSATS INSTITUTE OF INFORMATION TECHNOLOGY At 05:12 28 September 2014 (PT)
Does regional currency integration ameliorate global macroeconomic shocks in subSaharan Africa? The case of the 2008-2009 global financial crisis
Gregory N. Price Juliet U. Elu
Article information:
To cite this document:
Gregory N. Price Juliet U. Elu , (2014),"Does regional currency integration ameliorate global
macroeconomic shocks in sub-Saharan Africa? The case of the 2008-2009 global financial crisis", Journal
of Economic Studies, Vol. 41 Iss 5 pp. 737 - 750
Permanent link to this document:
http://dx.doi.org/10.1108/JES-08-2011-0092
Downloaded on: 28 September 2014, At: 05:12 (PT)
References: this document contains references to 49 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 22 times since 2014*
Access to this document was granted through an Emerald subscription provided by 543666 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com

Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0144-3585.htm
Does regional currency

integration ameliorate global
macroeconomic shocks in
sub-Saharan Africa? The case
of the 2008-2009 global
financial crisis
Global
macroeconomic
shocks
737
Gregory N. Price and Juliet U. Elu

Department of Economics, Morehouse College, Atlanta, Georgia, USA
Abstract
Purpose The purpose of this paper is to consider whether regional currency integration in subSaharan Africa ameliorates global macroeconomic shocks by considering the impact of the 2008-2009
global financial crisis on economic growth. This suggests that Central Africa Franc Zone (CFAZ)
eurocurrency union membership amplifies the effects of global business cycles in sub-Saharan Africa.
Design/methodology/approach The authors estimate the parameters of a quantity theory model
of economic growth within a Generalized Estimating Equation (GEE) Framework.
Findings Parameter estimates from GEE specifications reveal that the contraction in credit during
the financial crisis of 2008-2009 had larger adverse growth effects on sub-Saharan African countries
who were members of the CFAZ eurocurrency union. The authors also find that sub-Saharan African
countries who were members of the CFAZ eurocurrency union were more likely to experience
a contraction in credit.
Originality/value As far as the authors can discern, no existing empirical growth models
use a GEE framework to estimate parameters of interest. The GEE parameter estimates are
distribution-free, robust with respect to unknown forms of heteroskedasticity, and control for a wide
variety of error structures that can induce bias in panel data parameter estimates.
Keywords Financial crisis, Economic growth, sub-Saharan Africa, Currency unions,
CFAZ eurocurrency union, Contraction of credit
Paper type Research paper
I. Introduction
For developing sub-Saharan African countries, the increasing integration of their
capital markets with world capital markets one measure of so-called globalization
has costs and benefits (Senbet, 2001). The more(less) integrated sub-Saharan capital
markets are with that of world capital markets, the more(less) vulnerable are subSaharan African economies to global macroeconomic shocks that affect capital inflows.
Currency unions have been theorized to be an arrangement in which the conditions for
capital inflows are maximized (Alesina and Barro, 2002; Mundell, 1961), as they could
potentially promote exchange rate stability, which reduces the volatility of capital
inflows. In this context, currency union membership in sub-Saharan Africa has a
potential trade-off: higher capital inflows that can increase economic growth and raise
living standards vs increased vulnerability to adverse global macroeconomic shocks
JEL Classifications C23, F15, N17, O11, O47
Journal of Economic Studies

Vol. 41 No. 5, 2014
pp. 737-750
r Emerald Group Publishing Limited
0144-3585
DOI 10.1108/JES-08-2011-0092
JES
41,5
738
that can lower economic growth and living standards. The practical policy relevance of
these theoretical considerations about currency unions is important given the global
financial crisis of 2008-2009. The global financial crisis that started at the end of 2007
spread to sub-Saharan African countries, causing economic growth to fall 1.4
percentage points between by the end of 2008, and the International Monetary Fund
has lowered its post-2009 economic growth forecast for sub-Saharan African countries
(Brambila-Macias and Massa, 2010).
In this paper, we consider whether regional currency integration in sub-Saharan
Africa ameliorates global macroeconomic shocks by considering the impact of the
2008-2009 global financial crisis a time period generally viewed as encompassing
the crisis (Chor and Manova, 2012) on economic growth. In particular we consider
whether the 13 countries in sub-Saharan Africa that constitute the Central Africa Franc
Zone (CFAZ), relative to those countries in sub-Saharan Africa without functioning
regional currency union arrangements, were more or less vulnerable to the adverse
macroeconomic shock associated with the 2008-2009 global financial crisis[1]. If indeed
currency unions reduces exchange rate risk among members of a currency union, it can
condition the sensitivity of economies to adverse shocks in prices, and output (Boivin
et al., 2008). As such, relative to non-CFAZ countries in sub-Saharan Africa, the effect of
the global financial crisis on the growth path of CFAZ countries could be different.
As regional eurocurrency integration among countries conditions nominal output
on a common currency, we parameterize a parsimonious quantity theory model of
economic growth over the period 1960-2009 to capture effects of sub-Saharan African
country membership in CFAZ. In particular, we augment the quantity theory growth
model to account for membership in the CFAZ eurocurrency union and the contraction
of credit (Ziesemer, 2010) associated with the global financial crisis of 2008-2009. In our
view, a quantity theory model of economic growth is sensible as it is able to capture the
core conditions necessary for a currency union to be optimal. These core conditions
include monetary discipline, an optimal trade-off between the price level and output
stabilization (Fielding and Shields, 2001), and optimal debt to equity ratios or leverage
among firms (Bris and Koshkinen, 2002). Our analysis will inform the extent to which
currency union membership affects output about through financial de-leveraging, as
this can induce output changes through the money-credit relationship in a given
economy. As the quantity theory requires at least a long-run relationship between
nominal output and the nominal supply of money, the growth rate of nominal and real
living standards as measured by gross domestic product (GDP) are correlated across
time. Econometrically, we account for the possible time-dependent growth path of
prices, output and money supply by estimating the parameters of the quantity theory
model of nominal output growth within a semi-parametric Generalized Estimating
Equation (GEE) Framework that allows for a variety of possible error structures across
time.
Our inquiry is related to several strands of literature. As we consider the
consequences of the 2009 financial crisis on sub-Saharan African economies, our
analysis extends the recent policy-oriented contributions of Brambila-Macias and
Massa (2010), Fosu (2010), Fosu and Naude (2009), Ziesemer (2010), and Senbet (2001).
Our inquiry also contributes to the literature on the economic effects of currency
unions in sub-Saharan Africa as in Elu and Price (2010), Fielding and Shields (2001),
Debron et al. (2003, 2011), and Houssa (2008). As our theoretical framework is a
quantity theory model of economic growth, our analysis adds to the literature on the
empirical relevancy of the quantity theory in sub-Saharan African economies as in
Anoruo (2002), Fielding (1994), Henstridge (1999), Nwafor et al. (2007), Nwaobi (2002),
and Tabi and Ondoa (2011). Last but not least, our inquiry contributes to the literature
that considers the causes and consequences of economic growth in sub-Saharan
Africa[2]. Our particular inquiry is an extension of Elu and Price (2010), as our analysis
considers if regional currency integration in sub-Saharan Africa can ameliorate the
shocks to the growth of living standards that can occur in a global economy in which
sub-Saharan African countries are becoming increasingly integrated (Andrianaivo and
Yartey, 2010).
The remainder of this paper is organized as follows. In the second section, we
motivate a quantity theory model of economic growth as a specification for capturing
the cross-country effects of an adverse macroeconomic shock when there is variation in
currency union membership. The third section discusses the data and empirical
methodology. Given that our data are paneled across time, we adopt a GEE framework
to estimate the parameters in our growth model, which allows for a variety of plausible
time dependent error structures. We report parameter estimates in the fourth section.
The last section concludes.
II. A quantity theory model of economic growth
Our model of economic growth follows from how output evolves in an economy where
owners/managers of profit-maximizing firms can finance with credit the purchasing of
capital (K) and labor (L) services that determines an aggregate production function of
the form Q KaL1a. The typical firm owner/manager sells output at price p,
maximizes the utility of profit, and has a line of credit Mc which can be used to
purchase capital and labor services at cost r and w, respectively. For a linear utility
function with profit p pKaL1arKwL, the typical firms optimal input demand
is a solution to argmaxK,L [U(p) ( pKaL1arKwL)|Mc rK wL], where U(p) is
the utility of profit. Suppose the economy has a fractional reserve banking system
where a fraction of demand deposits are always lent, for a stock of money M this
establishes.
Proposition (optimal aggregate output and the supply of money)
In an economy where profit-maximizing firms can borrow from banks to finance the
purchase of capital and labor services, optimal aggregate output is increasing in the
supply of money.
A proof of this proposition follows from considering the typical profit-maximizing
firms optimal input demand which are K* (aMc)/r and L* [(1a)Mc]/w,
respectively. Substituting optimal input demands into the production function
yields Q* [aMc/r]a [((1a)Mc)/w]1a. If McAM, then McpM and qQ*/qM qQ*/
qMc [a/r]a [((1a))/w]1a40. Thus in an economy where owners/managers of firms
maximize the utility of profit and can borrow to finance capital and labor services,
optimal aggregate output increases with respect to increases in the supply of money.
Our theory suggests that the growth of output in an economy is determined by the
stock of money, which enable credit that banks lend to profit-maximizing firms. Thus
to estimate the effects of regional eurocurrency integration on economic growth in subSaharan Africa, we assume that for a typical country, nominal output evolves
according to the quantity theory as PQ MV, where P is the price level, Q is the real
level of output, M is the supply of money, and V is the velocity of money. If one assumes
that velocity is constant, the equation of exchange becomes a specification of the level
of nominal output or PQ b0M where b0 is constant velocity. Over time, the growth
Global
macroeconomic
shocks
739
JES
41,5
740
rate of nominal output in logs is simply logQ*tlogQ*t1 b0 logMtlogMt1, where

Q* PQ is nominal output and b0 a constant that can represent the constant value of
velocity over time[3]. This is a parsimonious specification that relates the growth in
nominal living standards in an economy to the growth rate of the money supply, and is
in the spirit of Friedman (1960). At its core a quantity theory model of economic growth
recognizes that the stock of money in the economy supports credit and loans that
profit-maximizing firms use to purchase capital and labor services to produce output.
III. Data and empirical methodology
To empirically determine whether regional currency integration in sub-Saharan Africa
ameliorates global macroeconomic shocks we use data from World Development
Indicators (2010) compiled by the International Bank for Reconstruction and
Development (World Bank)[4]. Our sample consists of 48 sub-Saharan African
countries over the 1960-2009 time period, which is a 49 year temporal window that in
our view enables identification of long-run steady state economic growth , and how it is
possibly perturbed by macroeconomic shocks such as the global financial crisis that
began in 2008. Our dependent variable from the WDI is nominal GDP in current
US dollars.
As a measure of a countrys stock of money, we use the current US dollar equivalent
value of liquid liabilities also known as M3 which is the sum of currency and deposits
in the central bank, plus transferable deposits and electronic currency, plus time and
savings deposits, foreign currency transferable deposits, certificates of deposit, and
securities repurchase agreements plus travelers checks, foreign currency time
deposits, commercial paper, and shares of mutual funds or market funds held by
residents. This definition of money supply is a broad measure of money available to
firms and households in an economy available through the financial/banking sector.
We create several binary variables to capture membership in the CFAZ
eurocurrency union, and the 2008-2009 global financial crisis. The binary variables
for a countrys currency union membership indicate whether or not a country is a
member of the CFAZ; West African Economic and Monetary Union (WAEMU); and
Economic and Monetary Union of Central Africa (CEMAC). The last two are simply
subsets of (1), and we disaggregate so as to determine if regional country groupings
with CFAZ are different as in Elu and Price (2010). Country observations for the years
2008-2009 are assigned a dummy variable to capture the commencement of the global
financial crisis, and whether or not there was a contraction in credit which is a
measure of the adverse macroeconomic shock associated with the global financial
crisis. The measure of credit is from the WDI, and defined as the financial resources
provided to the private sector, through loans, purchases of non-equity securities, and trade
credits and other accounts receivable, that establish a claim for repayment measured in
current US dollars.
Let qit and qit1 be current and previous period nominal GDP, respectively, for
country i at time t, we estimate the parameters of our quantity theory model of
economic growth based on regression specifications of the form:
lnqit lnqit1 b0 b1 lnmit lnmit1 b2 RCUit
b3 CRISISit b4 RCUit CRISISit eit
where [ln(mit)ln(mit1)] is the growth rate of the nominal money supply, RCUit is a
binary indicator of membership in the CFAZ eurocurrency union, CRISISit is a binary
indicator for credit contraction during the financial crisis of 2008-2009,

(RCUit CRISISit) is the interaction between the financial crisis and currency union
membership, and eit is a stochastic error[5].
As our sample constitutes repeated observations over time, this introduces the
possibility that for each country, annual growth rates of nominal GDP are correlated,
violating the assumption of independence required for identifying the parameters.
Moreover, the temporal nature of the possible correlations can introduce departures
from stationarity, further undermining parameter identification. Given this, we
estimate the parameters of the empirical growth specification in (1) within a GEE
framework (Zeger et al., 1988) that can accommodate alternative and plausible time
dependent correlation structures[6]. GEE parameter estimation is a semiparametric
approach where for k explanatory variables and j 1,2 y C possibly correlated
clusters of observations on the dependent variable Y i parameter estimates follow from
solutions to a quasi-score function of the form:
Uk b
C
X
Gi V1
i Yi ki 0
j1
where Gi ki/b, and for scale parameter / 1, Vi [(Ai)1/2Ri(a)(Ai)1/2]//. The Ai are

T T diagonal matrices with g(lit) as the tth diagonal element, a is a vector of
unknown parameters, and Ri(a) is the working correlation matrix.
Our choice of a GEE estimating framework instead of a standard likelihood-based
approach has an additional powerful motivation beyond the fact that GEE estimators
can accommodate within country correlation of the dependent variable, and departures
from stationarity in error structures. As Zorn GEE estimators are also a special case
of Generalized Method of Moments (GMM) estimators (Hansen, 1982). In particular,
GEE estimators are GMM estimators for which there are no overidentifying moment
restrictions. Given that GMM estimators do not require any distributional
assumptions, and are robust with respect to heteroskedasticity of unknown form,
our GEE estimation approach enables the estimation of parameters that more
convincing, as estimation proceeds from less restrictive parametric assumptions about
the probability distribution governing the dependent variable and expected value of
the residual error.
With respect to interpretation, GEE parameter estimates provide measures of
a population response to changing covariates. To the extent that assumptions about
the source of individual country heterogeneity under standard fixed effects estimation
is limited to whats available in the data, GEE parameter estimates provide population
average responses to covariates that are much closer to the data than parameter
estimates based on specifying the source of individual heterogeneity (Zeger et al., 1988).
In this context GEE parameter estimates will inform the extent to which the average
growth rates change when more multi-country currency unions form in sub-Saharan
Africa or the marginal average treatment effect.
IV. Results
Table I summarizes the covariates of our empirical specification of the quantity theory
model of economic growth. The number of observations reflects the number of
sub-Saharan African countries in the sample that had nonmissing values of all
the covariates under consideration over the 1960-2009 time period. Approximately
Global
macroeconomic
shocks
741
JES
41,5
Covariate
Description
Mean
SD
Number of
observations
ln(qit)ln(qit1)
Log difference in nominal gross domestic

product: (growth rate of nominal GDP)
Log difference in money supply: (growth rate of
money supply)
Currency union membership: binary variable
equal to one if member of currency union
Binary variable equal to one if there was a
contraction of credit during global financial crisis
0.062
0.144
693
0.082
0.176
693
0.365
0.482
693
0.003
0.054
693
ln(mit)ln(mit1)
742
Table I.
Covariate summary
RCUit
CRISISit
37 percent of the country observations are countries with membership in the

CFAZ eurocurrency union. Of these member countries, approximately three-tenths
of 1 percent experienced a contraction of credit during the 2008-2009 global financial
crisis. While this variation may seem too small to identify an effect, our GEE parameter
estimates can indeed exploit this variation to identify. As a GEE specification models
the average response to an independent covariate over a subpopulation that shares
common values of the covariate, parameter estimates account for responses to
particular independent covariates for the entire population some of which need not
have the same matchings of the dependent and independent covariate (Zeger et al.,
1988)[7].
GEE parameter estimates are estimated across seven different correlation
structures for the error term, are reported in Table II[8]. We allow for the correlation
r between growth rates to be exchangeable, independent, autoregressive of order 1
(AR1) and 2 (AR2), and stationary of order 1 (STA1) and 2 (STA2)[9]. The explanatory
adequacy of each regression is assessed with a Wald w2 distributed test for the null
hypothesis that exogenous explanatory variables have parameters that are jointly
insignificant. Given that GEE parameter estimates are not likelihood-based, both
relative and absolute goodness-of-fit measures based on likelihoods are not appropriate.
We report instead as an overall goodness-of-fit measure the Spearmans rank correlation
coefficient(rs) a nonparametric measure of the monotonic relationship between two
variates between the growth rate predicted from the GEE parameter estimates and the
actual observed growth rate[10]. To assess the best working correlation restrictions
across the specification, we report the Quasi-likelihood information criterion (QIC)
statistic of Pan (2001)[11]. In general, the specification with the smallest QIC statistic is
the best specification in terms of what GEE working correlation structure is most
consistent with the data.
Collectively, the GEE parameter estimates reported in Table II suggest a robust
effect of eurocurrency union membership on how sub-Saharan African economies
responded to the global macroeconomic shock associated with the 2008-2009 financial
crisis. In every GEE error specification, the sign on CRISISit and RCUit CRISISit is
negative and significant. This suggests that the 2008-2009 financial crisis significantly
lowered economic growth for sub-Saharan African countries in general, and for CFAZ
eurocurrency union members, economic growth was further reduced. The independent
effects of CFAZ eurocurrency membership is always insignificant. As for explanatory
power, the parameters are always jointly significant, and the quantity theory model
of growth appears to explain the data well, as the estimated rs of 0.7050 implies a
high concordance between actual observed growth and the growth predicted by the
Exchangeable
Independent
AR1
AR2
STA1
STA2
Notes: Standard errors in parentheses. *,**Significant at 0.01 and 0.05 levels, respectively
Regressand growth rate of nominal GDP (1960-2009)

Regressors
b0
0.013 (0.005)** 0.013 (0.005)** 0.013 (0.005)** 0.012 (0.005)* 0.012 (0.005)** 0.012 (0.005)**
ln(mit)ln(mit1)
0.619 (0.033)* 0.618 (0.033)* 0.627 (0.032)* 0.634 (0.031)* 0.627 (0.032)* 0.635 (0.031)*
RCUit
0.005 (0.006) 0.006 (0.006) 0.006 (0.006) 0.004 (0.006) 0.006 (0.006) 0.004 (0.006)
CRISISit
0.022 (0.008)** 0.032 (0.006)* 0.032 (0.006)* 0.031 (0.006)* 0.032 (0.006)* 0.031 (0.006)*
CRISISit RCUit
0.066 (0.013)* 0.055 (0.009)* 0.052 (0.010)* 0.054 (0.009)* 0.052 (0.010)* 0.054 (0.009)*
Diagnostics
Number of observations
693
693
693
689
693
689
w2k1 (H0: b1 ybk 0)
449.25*
555.47*
618.42*
646.06*
619.83*
646.86*
rs
0.7050
0.7050
0.7050
0.7050
0.7050
0.7050
Quasi-likelihood information criterion statistic
14.56
14.50
14.09
13.79
14.09
13.79
Error specification
Global
macroeconomic
shocks
743
Table II.
GEE parameter
estimates of quantity
theory growth model
JES
41,5
744
estimated growth model. In this context, the quantity theory model explains economic
growth in sub-Saharan Africa reasonably well[12].
The practical significance of the GEE parameter estimates can be assessed with a
consideration of the minimum QIC GEE parameter estimates the AR(2) and STA(2)
error structure specifications. Assume the average growth rate reported in Table I is an
approximation of the steady state growth rate for the period under consideration.
Evaluating the elasticity of growth with respect to the financial crisis reveals that the
financial crisis caused a reduction in economic growth relative to the steady growth
equal to approximately 50 and 135 percent, respectively, for non-CFAZ and CFAZ
eurocurrency union members[13]. This suggests that at least in the case of the 20082009 financial crisis, currency unions in sub-Saharan Africa have no benefits with
respect to ameliorating global macroeconomic shocks.
As our growth model captures the global macroeconomic shock associated with the
2008-2009 financial crisis through credit contraction, it leaves unanswered any insight
on causal pathways. In particular, what is particular about the CFAZ eurocurrency
union that render its members exposed, relative to non-members, to macroeconomic
shocks? Given our model, we speculate as a possibility that currency unions such as
CFAZ are more prone to credit contraction cycles. This is particularly likely to be true if
banks in countries with a common currency are more leveraged relative to banks in
countries that do not belong to currency unions. In their analysis of the leverage
decisions of exporting firms, Bris and Koshkinen (2002) find that common currency
unions have a tendency to induce leverage reductions among firms. To the extent that
over time, banks in currency union countries complement such leverage reductions,
this de-leveraging can cause credit contraction cycles which impact adversely on
economic growth. To explore this, Table III reports GEE probit parameter estimates,
where the dependent variable is whether or not a country had a contraction in credit in
a given year. The independent variable are membership in the CFAZ eurocurrency
union, and the liquid reserves to bank asset ratio of banks[14].
We estimate the GEE probit specifications with the identical error structures
utilized in the growth model parameter estimates, and report the same diagnostics.
Across all error specifications in the GEE probit parameter estimates in Table III reveal
that as the bank liquid reserves to asset ratio increases bank de-leveraging the
probability of credit contraction increases in all sub-Saharan countries. Moreover,
being a member of the CFAZ eurocurrency union has a positive and significant
effect on the probability of credit contraction. This suggests that banks in CFAZ
eurocurrency member countries are more prone to de-leveraging and credit contraction
cycles across time, and that the relatively high vulnerability of CFAZ countries to the
macroeconomic shock of the 2008 2008 global financial crisis is at least partially
explained by the relatively high propensity of banks in CFAZ eurocurrency union
countries to contract credit.
V. Conclusion
This paper considered whether regional currency integration in sub-Saharan Africa
ameliorates global macroeconomic shocks by considering the impact of the 2008-2009
global financial crisis on economic growth. We estimated the parameters of a quantity
theory model of economic growth augmented to account for a countrys membership in
the CFAZ eurocurrency union and the contraction of credit associated with negative
macroeconomic shocks. GEE parameter estimates reveal that the contraction in credit
during the financial crisis of 2008-2009 had larger adverse growth effects on member
0.681 (0.080)*
0.179 (0.091)***
0.821 (0.218)*
757
16.15*
934.89
757
16.11*
934.83
Independent
0.682 (0.081)*
0.178 (0.092)***
0.809 (0.213)*
Exchangeable
Notes: Standard errors in parentheses. *,***Significant at 0.01 and 0.10 levels, respectively
Regressand credit contraction (1960-2009)

Regressors
Constant
RCUit
Bank liquid reserves to asset ratio
Diagnostics
Number of observations
w2k1 (H0: b1 ybk 0)
Quasi-likelihood information criterion statistic
Error specification
757
18.21*
935.06
0.684 (0.082)*
0.177 (0.093)***
0.898 (0.216)*
AR1
757
15.80*
935.23
0.674 (0.083)*
0.171 (0.092)***
0.871 (0.232)*
AR2
757
16.25*
935.14
0.675 (0.082)*
0.172 (0.092)***
0.868 (0.227)*
0.687 (0.082)*
0.178 (0.093)***
0.907 (0.216)*
757
18.88*
935.05
STA2
STA1
Global
macroeconomic
shocks
745
Table III.
GEE probit parameter
estimates of credit
contraction
JES
41,5
746
countries. We also find that sub-Saharan African countries who were members of the
CFAZ eurocurrency union were more likely to experience a contraction in credit and
that the relatively high vulnerability of CFAZ countries to the macroeconomic shock of
the 2008 2008 global financial crisis is at least partially explained by the relatively
high propensity of banks in CFAZ eurocurrency union countries to contract credit This
suggests that CFA eurocurrency union membership amplifies the effects of global
business cycles in sub-Saharan Africa.
Our finding of a possible cost of currency integration in sub-Saharan Africa stand in
contrast to those of Anyanwu (2003), Balogun (2008), Elu and Price (2010), Geda and
Kebret (2007), Debron et al. (2003, 2011), and Masson (2008) which provide evidence
that common currency unions are, or can be beneficial for sub-Saharan African
countries. While we find a cost of eurocurrency union membership increased
vulnerability to adverse macroeconomic shocks such as the 2008-2009 financial crisis
our findings need not imply that currency unions in sub-Saharan Africa have no net
benefits. In this context, our results cannot inform the extent to which the CFAZ
eurocurrency union is optimal or not. Our parameter estimates do, however, suggest
that the cost incurred by CFAZ countries as a result of the recent 2008-2009 financial
crisis was substantial with respect to lowering the growth rate of nominal GDP relative
to its long-run steady state approximately 135 percent.
As we find evidence that the vulnerability of CFZ eurocurrency union members is at
least in part attributable to the propensity of the banking sector in CFAZ countries to
de-leverage and contract credit, our results have implications for financial firm
regulatory policy. Brownbridge and Kirkpatrick (2000) note that partly due to a
colonial legacy, inherited banking regulatory regimes in sub-Saharan Africa reflect
a colonial area banking environment in which there was little or no oversight by a
regulatory body. To the extent that this influences bank regulatory prerogatives and
credible oversight of risk in the banking system, our results suggest some scope for
better bank regulatory policy in CFAZ countries. As de-leveraging is a likely corrective
response to excessive leveraging which is correlated with risk, our results suggest that
bank regulatory policies that constrain bank leverage (e.g. more stringent capital
requirements) could reduce the vulnerability of CFAZ countries to adverse
macroeconomic shocks.
Notes
1. CFAZ consists of two separate regional currency unions in sub-Saharan Africa in which the
CFA franc currency is pegged to the euro: WAEMU and CEMAC. WAEMU includes the
countries of Benin, Burkina Faso, Cote dIvoire, Mali, Niger, Senegal and Togo. CEMAC
includes the countries of Cameron, Central African Republic, Chad, Congo Republic of,
Equatorial Guinea, and Gabon.
2. See, for example, Allen and Ndikumana (2000), Asiedu and Freeman (2009), Azam et al.
(2002), Bahmani-Oskoogee and Gelan (2007), Diop et al. (2010), Ekpo (1992), Elu and Price
(2010), Fosu (2008, 2009), Gyimah-Brempong and Corley (2005), Muhammad (2009), Osili
(2007), and Oyelere (2010).
3. Viewing the equation of exchange as a long-run equilibrium relationship (King and Watson,
1997), if the equilibrium velocity of money is stationary across time, b0 represents the
constant log value of velocity. For evidence that the velocity of money is stationary and hence
stable, see Herwartz and Reimers (2006).
4. These data are publicly available at: http://data.worldbank.org/
5. As we use the current US dollar equivalent of M3 as our measure of a countrys money

supply, its measurement in nominal terms captures in our regression specification captures
its effect on nominal output as our measures of GDP are in nominal terms. Given the
quantity theory that in the long-run the money supply only effects the price level, the
coefficient on b1 should be positive and significant, as increases in the nominal money
supply and nominal money should be positively correlated.
Global
macroeconomic
shocks
6. See Zorn for an overview of estimating GEE specifications when repeated observations data
are possibly correlated.
747
7. This virtue of GEE estimators also rules out distinct real versus nominal currency union
treatment effects. As our dependent variable is output in current US dollars, that GEE
parameter estimates do not require similarly measured covariates for identification of causal
effects suggests that our parameter estimates are not biased by any discrepancies between
real and nominal levels of output across the population of countries in our sample.
8. In contrast to Elu and Price (2010), our binary indicator for CFAZ eurocurrency union
membership does not distinguish between WEAMU countries and CEMAC countries, as our
GEE parameter estimates would not converge when considering WAEMU and CEMAC
countries separately.
9. In particular let Rjj0 be a working correlation matrix that includes for a given observation in
a country cluster the correlation in growth rates between year j and j0 for jaj0. Then for an
exchangeable process: Rjjt1 r, for an independent process: Rjjt1 I, where I is a T T
P t1
identity matrix, for an autoregressive process: Rjjt1 rjj j j , and for an order m

P
P
stationary process: Rjjt1 r j jt1 j if j jt1 pm, and zero otherwise.
10. As GEE specifications are nonlinear, rs is able to capture any monotonic relationship
between actual observed growth rates, and the predicted economic growth rate. In this
context the use of rs as a goodness-of-fit measure is similar in spirit to the concordance based
approach to assessing goodness-of-fit in GEE specifications proposed by Zheng (2000).
11. Cui (2007) provides an overview of QIC and how it is implemented in Stata.
12. Some caution could be exercised in interpreting our results as being consistent with the
quantity theory, as annualized observations, as opposed to multi-year averaged
observations, may not be an adequate time horizon for price-output adjustment. However,
our analysis is restricted by observations on the financial crisis, which commenced in 2009,
preventing multi-year averaging of it as an event.
13. This computation follows from simply estimating point elasticities based on:
qlnqit lnqit1 =qCRISISit 1=Slnqit lnqit1 T 1
where T is sample size, and 1 is the equilibrium value of the financial crisis as it is
dichotomous.
14. The liquid reserves to bank assets ratio is based on what is reported in the WID data. In our
sample, mean and standard deviation of this covariate is 0.145 and 0.172, respectively.
References
Alesina, A. and Barro, R.J. (2002), Currency unions, Quarterly Journal of Economics, Vol. 117
No. 2, pp. 409-436.
Allen, D.S. and Ndikumana, L. (2000), Financial intermediation and economic growth in
Southern Africa, Journal of African Economies, Vol. 9 No. 3, pp. 132-160.
Andrianaivo, M. and Yartey, C.A. (2010), Understanding the growth of African financial
markets, African Development Review, Vol. 22 No. 3, pp. 394-418.
JES
41,5
748
Anoruo, E. (2002), Stability of the Nigerian M2 money demand function in th SAP period,
Economics Bulletin, Vol. 14 No. 3, pp. 1-9.
Anyanwu, J.C. (2003), Estimating the macroeconomic effects of monetary unions: the case of
trade and output, African Development Review, Vol. 15 Nos 1/2, pp. 126-145.
Asiedu, E. and Freeman, J. (2009), The effect of corruption on investment growth: evidence from
firms in Latin America, Sub-Saharan Africa, and transition countries, Review of
Development Economics, Vol. 13 No. 2, pp 200-214.
Azam, J.-P., Fosu, A. and Ndungu, N.S. (2002), Explaining slow growth in Africa, African
Development Review, Vol. 14 No. 2, pp. 177-220.
Bahmani-Oskoogee, M. and Gelan, A. (2007), On the relation between nominal and real
devaluation: evidence from African countries, Journal of African Economies, Vol. 16 No. 2,
pp. 177-197.
Balogun, E.D. (2008), An alternative reconsideration of macroeconomic convergence criteria for West
African monetary zone, MPRA Working Paper No. 11367, University of Munich, Munich.
Boivin, J., Giannoni, M.P. and Mojon, B. (2008), How has the euro changed the monetary
transmission?, Working Paper No. 14190, National Bureau of Economic Research,
Cambridge, MA.
Brambila-Macias, J. and Massa, I. (2010), The global financial crisis and Sub-Saharan Africa:
the effects of slowing private capital flows on growth, African Development Review,
Vol. 22 No. 3, pp. 366-377.
Bris, A. and Koshkinen, Y. (2002), Coroporate leverage and currency crises, Journal of Financial
Economics, Vol. 63 No. 2, pp. 275-310.
Brownbridge, M. and Kirkpatrick, C. (2000), Financial regulation in developing countries,
Journal of Development Studies, Vol. 37 No. 1, pp. 1-24.
Chor, D. and Manova, K. (2012), Off the cliff and back? Credit conditions and international trade
during the global financial crisis, Journal of International Economics, Vol. 87 No. 1,
pp. 117-133.
Cui, J. (2007), QIC program and model selection in GEE analyses, Stata Journal, Vol. 7 No. 2,
pp. 209-220.
Debron, X., Masson, P.R. and Pattillo, C. (2003), West african currency union rationale and
sustainability, CESifo Economic Studies, Vol. 49 No. 3, pp. 381-413.
Debron, X., Masson, P.R. and Pattillo, C. (2011), Should African monetary unions be expanded?
An empirical investigation of the scope for monetary integration in Sub-Saharan Africa,
Journal of African Economies, Vol. 20 No. S2, pp. ii104-ii150.
Diop, A., Dufrenot, G. and Sanon, G. (2010), Is per capita growth in Africa hampered by poor
governance and weak institutions? An empirical study on the ECOWAS countries,
African Development Review, Vol. 22 No. 2, pp. 265-275.
Ekpo, A.H. (1992), Economic development under structural adjustment: evidence from selected
West African countries, Journal of Social Development in Africa, Vol. 7 No. 1, pp. 25-43.
Elu, J.U. and Price, G.N. (2010), The impact of the euro-CFA Franc Zone on economic growth in
Sub-Saharan Africa, Proceedings of the African Economics Conference 2008, Economica
Press, Paris.
Fielding, D. (1994), Money demand in four African countries, Journal of Economic Studies,
Vol. 21 No. 2, pp. 3-37.
Fielding, D. and Shields, K. (2001), Modelling macroeconomic shocks in the CFA Franc zone,
Journal of Development Economics, Vol. 66 No. 1, pp. 199-223.
Fosu, A.K. (2008), Democracy and growth in Africa: implications of increasing electoral
competitiveness, Economics Letters, Vol. 100 No. 3, pp. 442-444.
Fosu, A.K. (2009), Inequality and the impact of growth on poverty: comparative evidence for
Sub-Saharan Africa, Journal of Development Studies, Vol. 45 No. 5, pp. 726-745.
Fosu, A.K. (2010), The global financial crisis and development: whither Africa?, Working Paper
No. 2010/124, World Institute for Development Economics Research, United Nations
University, Helsinki.
Fosu, A.K. and Naude, W. (2009), The global financial crisis: towards syndrome-free recovery for
Africa, Working Paper No. 2009/03, World Institute for Development Economics
Research, United Nations University, Helsinki.
Friedman, M. (1960), A Program for Monetary Stability, Fordham University Press, New York, NY.
Geda, A. and Kebret, H. (2007), Regional economic integration in Africa: a review of problems
and prospects with a case study of COMESA, Journal of African Economies, Vol. 17 No. 3,
pp. 357-394.
Gyimah-Brempong, K. and Corley, M.E. (2005), Civil wars and economic growth in Sub-Saharan
Africa, Journal of African Economies, Vol. 14 No. 2, pp. 270-311.
Hansen, L.P. (1982), Large sample properties of generalized method of moments estimators,
Econometrica, Vol. 50 No. 4, pp. 1029-1054.
Henstridge, N.M. (1999), De-monetisation, inflation and coffee: the demand for money in
Uganda, Journal of African Economies, Vol. 8 No. 3, pp. 345-385.
Herwartz, H. and Reimers, H.-E. (2006), Long-run links among money, prices and output:
worldwide evidence, German Economic Review, Vol. 7 No. 1, pp. 65-86.
Houssa, R. (2008), Monetary union in West Africa and asymmetric shocks: a dynamic structural
factor model approach, Journal of Development Economics, Vol. 85 No. 1, pp. 319-347.
King, R.G. and Watson, M.W. (1997), Testing long-run neutrality, Economic Quarterly, Federal
Reserve Bank of Richmond, Vol. 83 No. 3, pp. 69-95.
Masson, P.R. (2008), Currency unions in West Africa: is the trade effect substantial enough to
justify their formation?, World Economy, Vol. 31 No. 4, pp. 533-547.
Muhammad, A. (2009), Would African countries benefit from the termination of Kenyas
economic partnership agreement (EPA) with the EU? An analysis of EU demand for
imported roses, Journal of Agricultural Economics, Vol. 60 No. 1, pp. 220-238.
Mundell, R.A. (1961), A theory of optimum currency areas, American Economic Review, Vol. 51
No. 4, pp. 657-665.
Nwafor, F., Nwakanma, H., Nkansah, P. and Thompson, F. (2007), The quantity theory of money
in a developing economy: empirical evidence from Nigeria, African Economic and
Business Review, Vol. 5 No. 1, pp. 1-9.
Nwaobi, G. (2002), A vector correction and nonnested modeling of money demand function in
Nigeria, Economics Bulletin, Vol. 3 No. 4, pp. 1-8.
Osili, O.U. (2007), Remittances and savings from international migration: theory and evidence
using a matched sample, Journal of Development Economics, Vol. 83 No. 2, pp. 446-465.
Oyelere, R.U. (2010), Africas education enigma? The Nigerian story, Journal of Development
Economics, Vol. 91 No. 1, pp. 128-139.
Pan, W. (2001), Akaikes information criterion in generalized estimating equations, Biometrics,
Vol. 57 No. 1, pp. 120-125.
Senbet, L.W. (2001), Global financial crisis: implications for Africa, Journal of African
Economies, Vol. 10 No. S1, pp. 104-140.
Tabi, H.N. and Ondoa, H.A. (2011), Inflation, money and economic growth in Cameroon,
International Journal of Financial Research, Vol. 2 No. 1, pp. 45-56.
World Development Indicators (2010), The International Bank for Reconstruction and
Development, World Development Indicators, Washington, DC.
Global
macroeconomic
shocks
749
JES
41,5
750
Zeger, S.L., Liang, K.-Y. and Albert, P.S. (1988), Models for longitudinal data: a generalized
estimating approach, Biometrics, Vol. 44 No. 4, pp. 1049-1060.
Zheng, B. (2000), Summarizing the goodness of fit of generalized linear models for longitudinal
data, Statistics in Medicine, Vol. 19 No. 10, pp. 1265-1275.
Ziesemer, T.H.W. (2010), The impact of the credit crisis on poor developing countries: growth,
worker remittances, accumulation and migration, Economic Modelling, Vol. 27 No. 5,
pp. 1230-1245.
Further reading
Zorn, C. (2006), Comparing GEE and robust standard errors for conditionally dependent data,
Political Research Quarterly, Vol. 59 No. 3, pp. 329-341.
Corresponding author
Dr Gregory N. Price can be contacted at: gnprice@langston.edu
To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints

Does Regional Currency Global Shock Ob Africa PDF

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Does Regional Currency Global Shock Ob Africa PDF

Uploaded by

Copyright:

Available Formats

Journal of Economic Studies

Downloaded by COMSATS INSTITUTE OF INFORMATION TECHNOLOGY At 05:12 28 September 2014 (PT)

About Emerald www.emeraldinsight.com

Downloaded by COMSATS INSTITUTE OF INFORMATION TECHNOLOGY At 05:12 28 September 2014 (PT)