Duplicative research, mergers and innovation

We show that in the model of Federico et al. (2017) horizontal mergers may actually spur innovation by preventing duplication of R&D efforts. Federico et al. do not notice this result because they presume that the merged firm spreads its R&D expenditure evenly across the research units of the merging firms-a strategy which is optimal, however, only if the probability of failure is log-convex in the RD effort. The possibility that mergers spur innovation is more likely, the greater is the value of innovations and the less rapidly diminishing are the returns to R&D.