Re-Evaluating the Knowledge Production Function for the Regions of the Russian Federation

The present study picks up on the aspect of knowledge generation—a key part of every national innovation system—in the context of the Russian Federation. Following Fritsch and Slavtchev (2006), a knowledge production function can be used to account for the efficiency of an innovation system. In detail, this study implements a panel quantile regression estimation approach and thus presents a novel approach in studying national innovation system and, more specifically, their efficiency. In particular, a non-linear knowledge production function is estimated to quantify for a possible non-linear impact of knowledge inputs on domestically—sing patents from the Russian Patent Office—and internationally—using patents from the European Patent Office—oriented knowledge output. Using regional data, it is shown that a non-linear impact of the inputs especially on Russian domestic patents can be found. The results offer new insights into the structure of the Russian innovation system as a threshold is identified where the innovation system switches from increasing returns of researcher input to decreasing returns. This implies that only smaller research systems work efficiently, and starting from a size of approximately 900 researchers, their efficiency steadily decreases.