Economic policy uncertainty : an index for Malta

We construct an Economic Policy Uncertainty (EPU) index for Malta based on the methodology of Baker, Bloom and Davis (2015) using articles published between 2004 and 2022 from four Maltese news portals. The index spikes during general elections and other significant local or international events. Using the 95% threshold zscore, we conclude that Malta experienced notably elevated EPU during four key events: Malta's accession to the European Union, the 2011/2012 and 2019 political crises, and the COVID-19 pandemic. To validate our index, we first analyse the behaviour of the time series by assessing the contents of the flagged articles through a bigram analysis, which shows that our index reflects the timeline of suggested instances where Malta experienced economic policy uncertainty. We use a vector auto-regressive model with exogenous variables to show that innovations to the EPU index exhibit countercyclical properties and reflect developments in the local economy beyond those attributable to negative sentiment. Furthermore, we highlight that, despite positive correlations, the behaviour of our index differs in timing and magnitude from foreign EPU indices, underscoring the need for a Maltese-specific index rather than relying on foreign-based proxies. Finally, the Toda and Yamamoto (1995) test supports the hypothesis that our EPU index Granger causes measures of household consumption at a monthly frequency, thus showing that our EPU index may be used as a leading indicator for this measure.