In Discussion Paper No. 1333, Research Fellow Nicholas Crafts surveys both the usefulness of endogenous innovation models of growth in economic history and the implications of historical research for new growth theorists. A distinction is drawn between models of endogenous growth that have capital accumulation as the engine of growth and those that emphasise technological change. The first part of the paper presents some descriptive data on growth in relatively rich economies since 1870. Evidence is also presented to suggest that the trend rate of growth has tended to vary substantially in most countries for which there is long-run data and to highlight the unusually fast growth of the early post-war years.

It is suggested that economic historians should take endogenous innovation models seriously and that this will help them to integrate traditional historians' emphasis on the impact of institutions and policy with cliometric research. A review of historical research suggests that growth economists should pay more attention to learning effects and technological shocks. Further research into measurement issues is shown to be seriously needed, notably, into establishing the magnitude and sources of total factor productivity growth.