Abstract
This paper investigates interindustry knowledge spillovers using input-output (IO) data and patent information, since spillovers can arise from different types of sources and linkages. Economic transactions of intermediate goods are used to track embodied knowledge flows with the data derived from Korean IO tables. In terms of disembodied knowledge flows, a technology proximity matrix is constructed using patent information. The empirical results show that patterns of interdependences among industries are different and this implies that policy-makers should take into account the interdependences between industries even when planning industry-specific policies. Moreover, excessive concentration of R&D investment needs to be improved.
Notes
1. PROXSCAL conducts multidimensional scaling of proximity data to find a least-squares representation of the objects in a low-dimensional space. For more details of the algorithm, the reader is referred to Commandeur and Heiser (Citation1993).