We study the one-year-ahead budgetary projections from the Stability and Convergence Programmes (SCPs) of EU Member States since the start of the Economic and Monetary Union (EMU) until the start of the coronavirus crisis. First, we consider errors of the general government’s headline budget balance, which we then split into expenditure and revenue errors. Next, we split the latter two into “base”, “growth” and “denominator” effects. We find that the most important explanatory variable is the GDP growth error: more optimism in GDP growth projections produces more optimistic budgetary projections. This effect goes beyond a mechanical denominator effect on spending and revenues as shares of GDP; it also works through the numerator of these ratios. Our findings may call for delegating the construction of output projections to adequately equipped national independent fiscal institutions. Finally, we explore how independent fiscal institutions shape projection errors. Those with high media impact producing or assessing the macroeconomic forecast appear to lead to actual budgetary improvement relative to projections.
Read full abstract