Suchergebnisse

AbstractThis article explores the robustness of Behavioural Equilibrium Exchange Rate models, employed to estimate real effective exchange rate misalignments, to the frequency of the underlying data. It compares misalignments stemming from an annual model, estimated since 1980, and a comparable quarterly model, estimated since 1999. The two sets of estimates are similar. Moreover, the in‐sample power of quarterly REER misalignments in explaining subsequent REER developments is higher than that of the annual estimates. This article therefore suggests that the "optimal" frequency of a BEER model depends on whether its resulting estimates are employed for research purposes or for policy‐making activities.

The study outlines the main challenges in measuring external imbalances via a "current account (CA) model", from which a CA "norm" may be derived and against which the actual CA is appraised. The selection of both the dependent and explanatory variables and technical issues in the estimation procedure are amongst the concerns raised. The impact on CA modelling of global trends, such as the expansion of MNEs, and of the COVID-19 pandemic is also discussed. The paper does not prescribe fully-fledged solutions, but rather provides awareness and hints on how to possibly tackle at least some of these challenges.