Does productivity increase with density? We revisit the issue using French wage and TFP data. To deal with the ‘endogenous quantity of labour bias (i.e., urban agglomeration is consequence of high local productivity rather than a cause), we take an instrumental variable approach and introduce a new set of geological instruments in addition to standard historical instruments. To deal with the ‘endogenous quality of labour bias (i.e., cities attract skilled workers so that the effects of skills and urban agglomeration are confounded), we take a worker fixed-effect approach with wage data. We find modest evidence about the endogenous quantity of labour bias and both sets of instruments give a similar answer. We find that the endogenous quality of labour bias is quantitatively more important.