Astrid Kander

Electricity has been regarded as a typical example of a general purpose technology and important for the surge both in energy productivity and overall productivity in the American economy in the 1920s. This view was challenged by Nicholas and Moser (2004) based on patent statistics. We argue that other methods are required for studying productivity effects and propose cointegration analyses. We demonstrate a clear impact from electrification on energy productivity in those broad Swedish industrial branches that used electricity for multiple uses. This effect goes beyond mere book-keeping effects and indicates the existence of dynamic effects.