Effects of Research Policy in Biotechnology: An Empirical Agent-Based Model of Knowledge Generation

Over the recent past, we can observe increasing interest in the ex-ante impact assessment of research policy, mainly related to the growing importance of accountability and limited budgets. However, existing methods often lack quantitative scenarios that go beyond extrapolations of current trends. This study addresses this research gap by proposing an empirical agent-based model (ABM) of knowledge generation in a system of researching firms. With our emphasis on the empirical calibration of ABMs, we intend to conduct scenario simulations applicable to real world contexts – in this study illustrated by means of an ABM on the Austrian biotechnology sector. In our model, effects of public research policy on the knowledge-related system output – measured by the patent portfolio – are under scrutiny. By this, the study contributes to the literature on ABMs in several aspects: Building on an existing concept of knowledge representation, we advance the model of individual and collective knowledge generation in firms by conceptualising policy intervention and corresponding output indicators. Furthermore, we go beyond symbolic ABMs of knowledge production by using empirical patent data as knowledge representations, adopt an elaborate empirical initialisation and calibration strategy using company data, and utilise econometric techniques to generate a sector-specific fitness function that determines the model output. With this model, we are able to conduct scenario analyses on effects of different public research funding schemes in the field of biotechnology. The results demonstrate that an empirically calibrated and transparent model design increases credibility and robustness of the ABM approach in the context of ex-ante impact assessment of public research policy.