Resolving the Anglo-German industrial productivity puzzle, 1895-1935: A response to professor Ritschl

This response offers a critical appraisal of the claim of Albrecht Ritschl to have found a possible resolution to what he calls the Anglo-German industrial productivity puzzle, which arose as the result of a new industrial production index produced in an earlier paper by the same author.(1) Projection back from a widely accepted 1935/36 benchmark using the Ritschl index showed German industrial labor productivity in 1907 substantially higher than in Britain. This presented a puzzle for at least two reasons. First, other comparative information from the pre-World War I period, such as wages, seems difficult to square with much higher German labor productivity at this time. 2 Second, a direct benchmark estimate produced by Stephen Broadberry and Carsten Burhop, using production census information for Britain and industrial survey material of similar quality for Germany, suggested broadly equal labor productivity in 1907.(3) Broadberry and Burhop also showed that if Walther Hoffmann's industrial output index was used instead of the Ritschl index for Germany, the puzzle largely disappeared(4). Ritschl now proposes some further changes to the German industrial output index, which move it closer to the Hoffmann index.(5) However, to remove the remaining discrepancy, Ritschl proposes a number of amendments to the 1907 benchmark(6). In this reply, we argue that most of the proposed changes to the benchmark are unwarranted. We also show that applying a uniform weighting scheme to the branch output series used in the new Ritschl industrial production index removes most of the remaining discrepancy.