Comment on “Stock markets and business cycle comovement in Germany before world war I: Evidence from spectral analysis”

This comment is focused mainly on the assumption that the HP filter is appropriate for isolating the cyclical component inherent in the time series investigated. It is argued that mechanical filtering can easily generate a spurious impression of cyclical behaviour. Estimating the cyclical component also requires analysis of the trend characteristics of the time series under investigation as well as the existence of irregularities in the data generating process. Different business cycle estimation methods are used to judge the robustness of the results for the NNP income series and the stock market index.