German national accounts for the 19th and early 20th Century

Abstract

The aim of this article is to review the national accounting figures Hoffmann et al. produced some time ago. Without presenting any reliable alternative time series the attempt is made to show the implausibilities and inconsistencies in Hoffmann sendeavour. This is done in two ways: first alternative plausible levels of relevant national accounting measures are calculated and discussed in order to demonstrate the direction of a specific bias and to indicate the consequences of certain assumptions (e.g. weighting schemes) in Hoffmann's figures. Secondly, I try to establish how Hoffmann S time series were in fact calculated. A preliminary conclusion suggests the construction of a new independent benchmark estimate for output before 1913. Should this diverge significantly from the Hoffmann-level a new attempt to reconstruct the German national accounts will be necessary.