ReviewThe impact of macroeconomic variables on exchange rate volatility in Ghana: The Partial Least Squares Structural Equation Modelling approach
Introduction
Exchange rate is defined as the value of a country’s currency in terms of another currency. In international trade, a country’s exchange rate could be used as the barometer of its international competitiveness. Consequently, volatility in exchange rate has serious far-reaching consequences for policymakers, investors, firms and consumers. For instance, exchange rate volatility negatively impacts investment decision making because it makes return on investment uncertain. Volatility of the exchange rate between the Ghana cedi and the international currencies especially the dollar has been a major concern to most Ghanaians especially entrepreneurs.
The adoption of the floating or flexible exchange rate system after the demise of the Bretton Woods institutions in 1973 has caused exchange rate fluctuations to be more violent than ever (Chang and Su, 2014). For instance, in the first quarter of 1986 after Ghana had adopted the managed exchange rate regime, about 0.01 Cedis exchanged for a dollar. However, by the end of April 2015, the rate had skyrocketed to 3.84, constituting a depreciation of 98.7% over the period (Adu et al., 2015). It is on record that the fluctuations in exchange rate could be influenced, in the short or long-run, by macroeconomic fundamentals, such as interest rate, output and price level. Consequently, the linkages between exchange rates and fundamentals are always given chary scrutiny by monetary authorities (Chang and Su, 2014). However, empirical studies on the relationship between exchange rate and macroeconomic fundamentals have yielded inconclusive results (Kwakye, 2015, Ray, 2008, Flood and Rose, 1999). Whereas some of the studies find a positive impact of macroeconomic fundamentals on exchange rate (Ray, 2008), some find that macroeconomic fundamentals play an insignificant role in understanding exchange rate volatility (Mark, 2009; Flood and Rose, 1999). Indeed, some studies argue that the relationship between macroeconomic variables and exchange rate is nonlinear (Yuan, 2011).
Pivotal to the debate on the determinants of exchange rate is the issue of the analytical technique adopted. The monetary model of exchange rates developed by MacDonald and Taylor (1994) applies the vector autoregressive (VAR) approach and utilizes the dollar-sterling exchange rate, as well as UK–US macroeconomic variables, in the period 1976–1990. The study concludes that the out-of-sample forecasting ability of an unrestricted monetary model achieves better results compared to the naïve random walk process. Mark (1995) conducts alternative bootstrap distributions under the null hypothesis that exchange rate is unpredictable. The study adopts a long-horizon regression and finds that monetary fundamentals have long-run forecasting ability to exchange rate changes. The Mark and Sul (2001)’s panel cointegration analysis of the interaction between nominal exchange rates and their fundamental values for a panel of 19 countries produces results to the effect that there is cointegration between exchange rates and fundamentals. However, the bivariate contingration tests performed by Engel and West (2005) fail to find cointegration between exchange rates and fundamentals. Similarly, Chang and Su (2014) analyse the dynamic relationship between exchange rates and macroeconomic variables data from Pacific Rim countries (monthly observations from 1986:01 to 2011:12 for Canada, Chile, Indonesia, Japan, S. Korea, Malaysia, the Philippines, Taiwan and the U.S., 1991:01 to 2012:07 for Singapore, and 1987:01 to 2006:12 for Thailand) and report that the conventional cointegration tests fail to find the long-run equilibrium for any country-pairs except Taiwan, but cointegration tests with structural breaks demonstrate the long-run connections between exchange rates and fundamentals for some country-pairs.
The above conflicting results have led some studies including Mark and Sul (2001) to posit that the elusive evidence of a long-run equilibrium relationship between exchange rates and fundamentals is attributable to the low power of conventional tests. Indeed, Evans and Lyons (2002) assert that the explanatory power of exchange rate determination models is essentially zero. Our motivation for this study is, thus, predicated on the argument that if the extant models are incompetent to establish any stable relationship between exchange rate and macroeconomic variables then the search for new models should go on unabated. It is our humble submission that the current study which applies the Partial Least Squares Structural Equation Modelling (PLS-SEM) approach to examine the effect of macroeconomic variables on exchange rate represents a contribution to the models for assessing the exchange rate-fundamentals nexus.
We apply the PLS-SEM approach to analyse data from Ghana which is one of the countries in Sub-Saharan Africa. Our choice of Ghana is influenced by two main reasons. First, Ghana is an import-dependent country with a tumbling currency relative to foreign currencies especially the dollar. Thus, unravelling the structural linkages between the cedi-dollar exchange rate and macroeconomic fundamentals should inform future policy interventions. Second, methodologically, it is our case that this study represents a marked departure from the extant literature in Ghana as far as the relationship between exchange rate and macroeconomic variables is concerned. The few studies that have been done with data from Ghana (e.g. Insah and Chiaraah, 2013; Mumuni and Owusu-Afriyie, 2004) have resorted to the use of regression and cointegration analytical methods, which lack the capacity to determine the complex structural relationship between macroeconomic variables and the cedi-dollar exchange rate.
We submit that the current study is relevant for some reasons. Exchange rate stability achieved through stable macroeconomic variables gives credibility to the economy and communicate very good news to both local and foreign trade partners. Fluctuations in exchange rates may have a direct influence on price stability, financial stability as well as trade balance (Ozkan and Erden, 2015). Importantly, the adoption of inflation targeting strategy by many central banks has accentuated the need to monitor the connection between exchange rates and macroeconomic variables. The findings of the study provide fresh insights into the relationship between exchange rate and macroeconomic fundamentals which will immensely assist central banks including the Bank of Ghana to monitor the macroeconomic variables that impact exchange rate. A deeper understanding of the complex structural relation between macroeconomic variables exchange rate provides insights to policy makers in budgeting for the various sectors of the economy to ensure its (the economy’s) efficient performance.
Section snippets
Selection of variables
Based on the extant literature (Kwakye, 2015, Sani and Idakwoji, 2014, Abbas et al., 2012, Adom et al., 2012, Kim and Roubini, 2000) we assess the impact of inflation (INF), monetary policy rate (MPR); current account balance (CAB), money and quasi money supply per GDP (M2GDP), annual GDP growth rate (GDPGRO), and total external debt (EDBTOT) on the Cedi-dollar exchange rate (EXRATE).
EXRATE is the dependent variable in our model. INFL, MPR, CAB, M2GDP, GDPGRO and EDBTOT are the independent
NEGCORR” latent variable
MPR, INF and CAB have inverse relationships with EXRATE. It implies that an increase in MPR, INF and CAB is likely to trigger the depreciation of the Ghana cedi against the dollar. Fig. 10 illustrates the trend of these indicators with the cedi/dollar exchange rate.
“POSCORR” latent variable
GDPGRO, M2GDP and EDBTOT correlate positively with EXRATE. Fig. 11 depicts the relationships between the three macroeconomic variables and EXRATE.
Convergent validity
The combined loadings and cross-loadings are used to describe convergent validity of
References (33)
- et al.
The dynamic relationship between exchange rates and macroeconomic fundamentals: Evidence from Pacific Rim countries
J. Int. Financ. Mark. Inst. Money
(2014) - et al.
Exchange rate anomalies in the industrial countries: a solution with a structural VAR approach
J. Monet. Econ.
(2000) - et al.
The monetary model of the exchange rate: long-run relationships, short-run dynamics andhow to beat a random walk
J. Int. Mon. Financ.
(1994) - et al.
Nominal exchange rates and monetary fundamentals: evidence from a small post-Bretton woods panel
J. Int. Econ.
(2001) The exchange rate and macroeconomic determinants: Time-varying transitional dynamics
North Am. J. Econ. Finan.
(2011)- et al.
Relation between GDP, inflation and real interest rate with exchange rate fluctuations of african countries
Int. J. Acad. Res. Account. Finance Manag. Sci.
(2012) - et al.
Sources of real exchange rate volatility in Africa. The case of Ecowas
Afr. Dev. Rev.
(2012) - et al.
An Empirical Analysis of Exchange Rate Dynamics and Pass-Through Effects on Domestic Prices in Ghana
(2015) - et al.
Interpretation of formative measurement in information systems research
MIS Q.
(2009) Statistical Power Analysis for the Behavioural Sciences
(1988)
Exchange rates and fundamentals
J. Polit. Econ.
Order flow and exchange rate dynamics
J. Polit. Econ.
Understanding exchange rate volatility without the contrivance of macroeconomics
Econ. J.
Evaluating structural equation models with unobservable variables and measurement error
J. Mark. Res.
A predictive approach to the random effects model
Biometrika
Multivariate Data Analysis
Cited by (14)
Acceptance of digital investment solutions: The case of robo advisory in Germany
2021, Research in International Business and FinanceCitation Excerpt :With a maximum of 7 exogenous constructs pointing towards usage intention in our baseline model (see Fig. 1), our sample meets this criterion. PLS recently gained increasing acceptance in banking and finance research with applications including construction of sentiment indices (Huang et al., 2015; Xu and Zhou, 2018), the assessment of the impact of macroeconomic variables on exchange rate volatility (Adusei and Gyapong, 2017), prediction of financial distress (Mselmi et al., 2017), transmission of systemic risk and stress testing (Avkiran et al., 2018), and the analysis of credit relationships between savings banks and small and medium-sized enterprises (SMEs) (Hirsch et al., 2018).6 As a last robustness check, we follow Avkiran (2018) and Avkiran et al. (2018) and re-estimate our model using generalized structured component analysis (GSCA)8 introduced by Hwang and Takane (2004, 2014).
Foreign currency exchange rate prediction using non-linear Schrödinger equations with economic fundamental parameters
2021, Chaos, Solitons and FractalsCitation Excerpt :The behavior of the exchange rate is important to understand because it is needed to formulate a country's policies that aim to achieve macroeconomic stability in an economy. Uncertainty in the behavior of the exchange rate will disrupt the macroeconomic stability that has been determined by a country [1]. It has become an interesting challenge to research international financial issues, namely to develop a reliable prediction model of the out-of-sample exchange rate using economic fundamentals.
A bird's eye view of Ghana's renewable energy sector environment: A Multi-Criteria Decision-Making approach
2021, Utilities PolicyCitation Excerpt :Therefore, volatility in the exchange rate in a country has a significant impact on the investor community, policymakers, consumers, and businesses. Volatility in the Ghanaian cedi exchange rate between other international currencies, particularly the United States dollar, remains a significant concern to most Ghanaians, primarily the business community (Adusei and Gyapong, 2017). The use of the flexible or floating type of exchange rates after the 1973 collapse of the Bretton Woods Institutions resulted in fluctuations in exchange rates more violently than ever (Adusei and Gyapong, 2017; Chang and Su, 2014).
RMB exchange rate forecasting using machine learning methods: Can multimodel select powerful predictors?
2024, Journal of ForecastingEconomic growth through global value chains; new insight from exchange rate effects on the African economy
2024, Cogent Economics and FinanceCurrency reform, currency biases and Ghana’s forex market fluctuations: Beyond the macroeconomic fundamentals
2023, Cogent Economics and Finance