The UK (and Western) productivity puzzle: does Arthur Lewis hold the key?

I propose a new explanation for the UK productivity puzzle. I graft the Lewis (1954) model onto a standard Solow growth model. What I call the neo-Lewis model is identical to the Solow model in good times. But in bad times foreign demand for a country’s exports is constrained below potential supply. This makes labour productivity growth depend negatively on the growth of labour input. I also argue that the neo-Lewis model can explain the fall in TFP growth, in the UK and elsewhere, after 2007. The predictions of the neoLewis model are tested on data for 23 advanced countries and also on a larger sample of 52 countries and find support.