Economic growth and innovation complexity: An empirical estimation of a Hidden Markov Model

Over the past decades, research effort in high income countries has substantially increased. Meanwhile, the growth rates of per capita output have been rather stable. The contribution of this paper is twofold. The first is to provide a theoretical explanation for such trends by developing an R&D-based growth model which accounts for dilution, difficulty and duplication effects. The second is to show empirically that the occurrence of different phases in the economic growth dynamics traces back to the interplay between complexity and specialization in production. To do this we estimate a Hidden Markov Model in which countries can switch across different growth regimes. We identify four distinct growth regimes.