Optimal lag when analysing the fiscal sustainability through the cointegration approach

The present Dissertation studies the fiscal sustainability in European Union countries by analyzing the relationship between government revenue and expenditure, and the selection of the optimal lag length. Using annual data from 1995 to 2024 for the 27 European Union countries, vector autoregressive (VAR) models and Granger causality tests are applied. The optimal lag for each country is selected based on standard information criteria (AIC, HQIC, and SBIC). The results show that, in many countries, revenue Granger-causes expenditure; causality in the opposite direction is less frequent. Regarding optimal lags, it varies considerable across country – in this study the range considered was one to ten. These findings highlight the importance of considering lag structure when examining fiscal sustainability.