Impact of Shocks on the Balance of Payment Evidence via MGARCH from Sudan

Abstract

This research investigates the effects of own shocks on the Sudanese balance of payment (BOP) and shocks from other variables. The model encompasses seven variables that have long-term relationship with each other i.e. ratio of balance of payments to GDP (BOPR), economic growth proxied by per capita GDP (Q), real effective exchange rate (REER), budget deficit ratio to GDP (BUDR), monetization (MNT), inflation rate (INF), and unemployment rate (UR). We use a multivariate generalized autoregressive conditional heteroscedasticity model (MGARVH) for estimation. We chose t distribution because residuals are not multivariate normal. All estimates are highly significant and positive. Positive ARCH effects suggest that a shock to the ith variable has positive effects on all variable covariances in the next period. The multiplication of own shocks by the cross shocks discloses that the covariance of the balance of payments and the inflation rate is the largest contrary to the smallest amount covariance of the balance of payments and the fiscal deficit. The positive estimates of the cross terms capture linkages, the transformation of shocks and volatility spillover effects point toward model variables integration. Covariance stationary test indicates multidirectional volatility spillover runs from single variable to the other variables. Dynamic conditional correlation (DCC) reveals a strong correlation of the variable model comprising both signs. We recommend that policymakers when formulating macroeconomic policies should consider the linkages between these variables taking Kaldor magic square as a guideline.