Abstract

This empirical study explores how news events affect the exchange-rate return volatility of the USD/AFN and whether positive and negative news shocks have asymmetric effects on the exchange-rate volatility. In this study, the daily exchange rate data for USD/AFN are taken from the website of Da Afghanistan bank, from January 2023 to May 2026, and are 1034 observations. There are six types of news events: aid inflows, aid suspension, economic, political, migration, and monetary. Second, aggregated positive and negative news variables are also created to look at the asymmetric effects of news shocks. The empirical analysis uses Autoregressive Moving Average and Autoregressive Conditional Heteroscedasticity models, to reflect the dynamic properties and the time-varying volatility of the series of exchange rates. The results suggest volatility clustering of the USD/AFN exchange rate, which implies that volatility levels aren't randomly distributed across time, but instead volatility extremes tend to stick around, while past information helps explain volatility levels today and predict volatility in the future. The findings also indicate that the news is more likely to influence the volatility of exchange rates than the returns. The news categories analyzed reveal that aid suspension, economic, political and monetary news have a significant impact on exchange-rate volatility, while migration news has no significant impact at all. The analysis also has evidence of the asymmetric reactions to news shocks, in the sense that negative news shocks cause much more volatility than positive news shocks. The results underscore the need to monitor news-related shocks and enhance information management in the foreign-exchange market in Afghanistan and give relevant inputs to the policymakers, Da Afghanistan Bank, financial institutions, and foreign-exchange market players for assessing and managing exchange-rate risk.

Keywords

Afghanistan GARCH Model Exchange Rate Volatility News Impact USD/AFN