The impact of political instability on exchange rates in South Africa: an econometric modelling
- Authors: Munzhelele, Tshilidzi Whitney
- Date: 2023-12
- Subjects: Econometric models , Economics -- Statistical methods , Finance -- Econometric models , Foreign exchange rates -- Econometric models
- Language: English
- Type: Doctoral theses , text
- Identifier: http://hdl.handle.net/10948/65861 , vital:74278
- Description: The exchange rate and political instability are crucial macroeconomic variables strongly related to every economy. In South Africa, exchange rate fluctuations are widely related to major political events. The study analyses the fluctuations in exchange rates by applying exchange rate data over the period 1989 to 2020. The current study, used the predictive quantitative design that combines correlational analysis with predictive modelling. The Unit root results show that political instability and exchange rate are stationary at first difference, and inflation, GDP, political instability, the rule of law, and corruption control and corruption freedom are stationary at level. The Vector Auto Regression model (VAR) was applied to examine the short-run relationship between political instability exchange rate, GDP, inflation, real interest rate, corruption and the rule of law. The results show a short-run relationship between political instability, exchange rate, GDP, inflation, real interest rate, corruption and rule of law. Johansen Cointegration testing was also performed to establish cointegration between variables. The results from the Johansen tests suggest that the model presents a cointegration between political instability, GDP, inflation, real interest rate, corruption and the rule of law and exchange rate, implying that these variables are related and can be combined linearly. The VECM was performed to establish a long-run relationship between variables since cointegration was established between variables. The Vector Error Correction (VEC) model complemented these findings resulting in the null hypothesis that states there is no long-run relationship between variables being rejected. The alternative hypothesis that there is a long-run relationship between variables was accepted. The Granger Causality test was performed to examine the causality between variables and to examine the drivers and causes of exchange rate fluctuations in the VAR model. The results revealed that political instability does not Granger cause exchange rate fluctuations in the short run and that there is a negative relationship between political instability and exchange rate fluctuations. The short-run results revealed that the exchange rate does not Granger cause political instability. However, the exchange rate Granger causes the country's political instability in the long run. This study’s literature review found that political instability harms exchange rates and the economy, and its impact can be felt globally. The results of the study show that there is a negative relationship between political instability and exchange rates. In the short run, the results show that political instability Granger causes inflation. The impulse response function (IRF) was conducted to determine the shock of political instability on the exchange rate. The findings indicate that the magnitude of the shock refers to one standard deviation. The results show that after two years, 92% of the shock in exchange rates is due to the shock on the exchange rate itself, and only 0.12% is due to political instability in the short run. This means that the shock on the exchange rate is associated with the exchange rate itself. The result is consistent with empirical findings in South Africa that fluctuations, in the long run, are largely caused by political instability from corruption. The exchange rate results and the political instability response were used to calibrate the long-run response to exchange rate fluctuation. Political instability was also used to assess the nexus between political instability and economic growth. The results determine a positive relationship between political instability and economic growth. The IRF was performed to track the impact of a variable on other variables (that is, the exchange rate on political instability, GDP, corruption, inflation, real interest rate, the rule of law, and corruption) in the system from several periods in the future. The Autoregressive Integrated Moving Average (ARIMA) model for forecasting was selected because it provides an accurate forecast and satisfies the criteria for an ideal model. The results show that in the future, not much variation can be expected in the long run, meaning that political instability is projected to stabilise from 2021 to 2040, and the exchange rate will increase. The results from the analysis of exchange rates and political instability confirm the existence of a negative relationship between political instability and exchange rates. The findings of the study point to a need for the South African government to immediately respond to an increase in exchange rates and to stabilise the undercurrents caused by macroeconomic shocks. The study will contribute to the theoretical understanding of fluctuations in exchange rates and the formulation of macroeconomic stabilisation. , Thesis (PhD) -- Faculty of Business and Economic Sciences, School of Economics, Development and Tourism , 2023
- Full Text:
- Date Issued: 2023-12
Model selection for cointegrated relationships in small samples
- Authors: He, Wei
- Date: 2008
- Subjects: Economics -- Statistical methods , Cointegration -- South Africa , Econometrics
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:10570 , http://hdl.handle.net/10948/971 , Economics -- Statistical methods , Cointegration -- South Africa , Econometrics
- Description: Vector autoregression models have become widely used research tools in the analysis of macroeconomic time series. Cointegrated techniques are an essential part of empirical macroeconomic research. They infer causal long-run relationships between nonstationary variables. In this study, six information criteria were reviewed and compared. The methods focused on determining the optimum information criteria for detecting the correct lag structure of a two-variable cointegrated process.
- Full Text:
- Date Issued: 2008