Shocks effects of inflation, money supply, and exchange rate on the West African Monetary Zone (WAMZ): Asymmetric SVAR modelling

Abstract

This paper has investigated the shocks effects of inflation, money supply, and exchange rate on the economies of the West African Monetary Zone (WAMZ) from 1987 to 2019 using the Kapetanios-Shin-Snell nonlinear cointegration test, Kilian-Vigfusson asymmetric tests, and Hatemi technique that allows the estimation of Asymmetric Structural Vector Autoregressive (ASVAR) model. The findings revealed that in all the countries, the shocks effects of inflation, money supply, and exchange rate are asymmetric except in Guinea and inflation in Liberia. Furthermore, for Gambia and Nigeria, only money supply is impacting the economies, while for Ghana, Guinea, and Liberia, none of the variables is impacting the economies but for Serra Leone, money supply and exchange rate are impacting its economy. Moreover, all the countries have common sources of shocks emanating from monetary and exchange rate policies except Gambia, which is monetary policy only. Therefore, the paper recommends the members, especially Ghana, Guinea, and Liberia, a solemn effort on appropriate monetary and exchange rate policies to boost their economies. Also, since almost all the countries have common sources of shocks emanating from monetary and exchange rate policies, they can embark into the monetary integration.

KEYWORDS:

• Inflation
• money supply
• exchange rate
• Asymmetric SVAR modelling
• shocks effects
• West African Monetary Zone

JEL Classifications:

• E31
• F33
• F41

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are openly available in the World Bank Development Indicator at https://data.worldbank.org/. Nonetheless, we have incorporated the copy file of the data used in the repository of the submission system.

Notes

1 Note, it is required that at least (n²–n) ÷ 2 restrictions be placed for the model to be just identified where n is the number of variables in the model. Hence, the number of identification should at least = (4²–4) ÷ 2 = 6. Looking at matric A, it can be seen that the model is exactly identified since the number of identification in the model is also 6.