Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorGarfagnini, Umbertoen_US
dc.contributor.authorStrulovici, Brunoen_US
dc.description.abstractWe study social learning and innovation in an overlapping generations model, emphasizing the trade-off between marginal innovation (combining existing technologies) and radical innovation (breaking new ground). We characterize both short-term and long-term dynamics of innovation, and the intergenerational accumulation of knowledge. Innovation cycles emerge endogenously, but the number of cycles is finite almost surely, and radical innovation terminates in finite time. We identify a negative relationship between past successes and the magnitude of radical innovation, combining insights from the multi-armed bandit literature with a spatial representation of innovation. Past successes reduce the incremental value of experimentation, and result in less ambitious innovation. In our framework, patents promote radical innovation through two channels: by increasing the expected benefit of radical innovation and by increasing the cost of marginal innovation. Our analysis suggests that sustaining radical innovation in the long-run requires external intervention.en_US
dc.publisher|aNorthwestern Univ., Kellogg Graduate School of Management, Center for Mathematical Studies in Economics and Management Science |cEvanstonen_US
dc.relation.ispartofseries|aDiscussion Paper, Center for Mathematical Studies in Economics and Management Science |x1546en_US
dc.subject.keywordSocial Learningen_US
dc.subject.keywordTechnological Changeen_US
dc.titleSocial learning and innovation cyclesen_US
dc.typeWorking Paperen_US

Files in This Item:
646.09 kB

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.