A method is developed to empirically test the hypothesis of induced innovation as it has been specified and used in the theoretical literature. A strong and a weak version of the hypothesis is tested using sectorial data from the USA, Canada, Germany, France and the UK. The strong version tests for the exact dependency of the relation between the change in factor-productivities on the one hand and relative prices and actual factor-productivities on the other hand. The weak version only tests for the direction of this dependency. In all countries the weak hypothesis is accepted in all sectors except in 'electricity, gas, and water'. The strong hypothesis is accepted in about half of all sectors. It is rejected only in sectors, in which the degree to which progress is intentional is low.