A New Empirical Index to Track the Technological Novelty of Inventions: A Sector Level Analysis

We introduce the Knowledge Origin Re-Combination Index (KORCI) to measure the ex-ante technological novelty of inventions at the sectoral level. The index is developed through the intertemporal comparison of a sequence of networks, which represents the complex connections between the technological components listed in subsequent cohorts of patent applications. This allows us to quantify the intensity of recombination of components and the introduction of new ones at the frontier of technological knowledge. Using patent data from three sectors - artificial intelligence, computer technology, and pharmaceuticals - we are the first to document the cyclical nature of the evolution of ex-ante technological novelty of inventions across all three sectors. These evolutionary cycles, however, are not synchronized and therefore it is unlikely that they are driven by a common innovation engine. Further investigation into the correlation between KORCI and patent growth rates reveal other differences among the sectors in both direction and strength. We conjecture that the relation between the degree of ex-ante technological novelty and invention activities depends on the specific innovation environment of the sector – whether these are process-based or product-based. Our new tool opens opportunities for new empirical research into the evolution of innovation at the sectoral level.