Explore publications.

The Marshall-Lerner condition states that a trade balance improves after currency depreciation if trade elasticities exceed one. Using the Vector Error Correction Method (VECM), this study tests the condition for South Africa, focusing on how exchange rate changes affect the trade balance under a floating exchange rate regime. The analysis examines short- and long-run impacts on exports and imports to assess whether exchange rate adjustments can improve the trade balance. Findings reveal that global income growth and exchange rate appreciation positively influence exports, while imports increase with both appreciation and higher domestic income. Although exports and imports respond strongly to exchange rate changes, the Marshall-Lerner condition does not hold for South Africa, suggesting that exchange rate adjustments alone may not improve the trade balance.

Several official dating institutions viz: SARB, NBER, OECD, CEPR, and others, provide Business Cycle (BC) chronologies with lags ranging from three months to several years. This is problematic for the various oppressed as it poses limits to the usefulness of BC indicators in terms of forecasting and policy implementation at large. In this article, we propose the construction of composite indicators as accurate measures of business cycles in South Africa. We use this composite index to facilitate comparison of both a non-parametric Bry and Boschan dating algorithm and a parametric Markov Switching dating method, in terms of performance and accuracy of dating business cycle turning points in South Africa. Utilising data spanning the period 2000M1 – 2018M12, empirical evidence obtained is such that, composite indices are better than single indicators due to information rich. Further, while the parametric and non-parametric methods’ performances are matched in terms of the number of turning points identified, however, the non-parametric method is more accurate in identifying these turning points. Due to its accuracy the non-parametric method proved to be a promising method of dating business cycle turning points in South Africa and alleviate the problems which are currently facing the South African Reserve Bank, the private sector and the economy at large.

The article reports on the complementarity and substitutability of Macroprudential and Monetary policies in South Africa, based on the interactions of both Business and Financial cycles. To this end, a Dynamic Conditional Correlations and the Asymmetric Dynamic Conditional Correlations MGARCH models together with the Granger Causality model, the Artificial Neural Network VAR model, and the Structural VAR model, were adopted for the analyses of synchronization, causality, and the analysis of shocks to cycles, respectively. Empirical evidence obtained is such that, under conditions of financial and real economic stress in South Africa, there exists high synchronicity and bidirectional causality between CBCI and CFCI. Hence, Macroprudential and Monetary policies become complements, there exists interdependence between the two policies and actions of one policy contributes to the improvement of the other. However, under normal times there exist no synchronicity and a unidirectional causality relationship running from CBCI to CFCI, was observed. Under such conditions, Macroprudential and Monetary policies become substitutes, therefore, the two policies become independent of each other, hence, only one policy can achieve the desired outcome. Overall, a shock to one cycle is a major determinant of the fluctuations of the other cycle.

This study aimed to investigate economic growth (RGDP) and its determinants such as internal debt (INDT), external debt (EXDT), interest rate (RINR), exchange rate (REXR) and trade openness (OPEN). Quarterly data from 1986-2021 used for this investigation were extracted from Central Bank of Nigeria.  Exploratory data analysis (EDA) revealed linear dependence of RGDP and the aforementioned determinants. The EDA and the variance inflation factor revealed the presence of multicollinearity caused by the INDT. As a result of this, the ridge regression method was used and it was found that the multicollinearity problem was addressed with appropriate ridge constant K= 0.29. Hence, ridge regression technique with constant K= 0.29 was a robust method to predict economic growth in Nigeria using the explanatory variables under consideration. Thus, served as a great benefit to the policy makers as the study provided a better understanding of the economic growth and the aforementioned determinants in terms of relationship and in particular the set back caused by EXDT on the economy of Nigeria. Also, the researchers benefited from this study as it engendered the understanding of the appropriate estimation technique to be used when the presence of multicollinearity was established in fitting a linear model.