Bounded Learning from Incumbent Firms

Abstract

Social learning plays an important role in models of productivity dispersion and longrun growth. In economies with a continuum of producers and unbounded productivity distributions, social learning can sometimes leave long-run growth rates completely indeterminate. This paper modifies a model in which potential entrants attempt to imitate randomly selected incumbent firms by introducing an upper bound on how much entrants can learn from incumbents. When this upper bound is taken to infinity, a unique long-run growth rate emerges, even though the economy without upper bound has an unbounded continuum of balanced growth rates.

Reconciling Models of Diffusion and Innovation: A Theory of the Productivity Distribution and Technology Frontier

“…In our paper, we explore how adoption and innovation choices respond to economic incentives and generate endogenous growth. Luttmer (2020) diagnosed the issue of multiple equilibria and hysteresis in idea‐diffusion models and showed how the introduction of a finite maximum growth rate for agents that learn from technology diffusion can generate uniqueness.…”

“…77 Luttmer (2012a, 2015a, 2015b, 2020) provided careful analysis of the role of hysteresis, including the important interaction of the stochastic innovation process with initial conditions. …”