National innovation systems, economic complexity, and economic growth : country panel analysis using the US patent data

This study examines the impacts of national innovation systems (NIS) and economic complexity index (ECI) on economic growth. A composite index of NIS is developed by using US patent data as a weighted sum of three, four or five variables among the following: concentration of assignees, localization, originality, diversification, and cycle time of technologies. Growth regressions confirm the significant and robust impacts of NIS3a, NIS4a, and NIS5 indices on economic growth. The common feature of these NIS indices is that they have the same component variables as their ingredients, and these are originality, cycle time, and technological diversification. NIS3a is the most parsimonious and powerful among all indices. The robustness of ECI is questionable because ECI loses significance after adding government expenditure and terms of trade variables into the regression model. Results confirm the overall importance of NIS in economic growth and justify policy efforts to improve NIS. This research is one of the first to generate a robust NIS index by using patent data only without many data requirements and free from the problem of cross-country comparability of underlying variables.