Reinganum (1989) - The Timing Of Innovation Research Development And Diffusion

Jump to navigation Jump to search
Has bibtex key
Has article title The Timing Of Innovation Research Development And Diffusion
Has author Reinganum
Has year 1989
In journal
In volume
In number
Has pages
Has publisher
©, 2016


  • Reinganum, Jennifer F. (1989), "Chapter 14 The timing of innovation: Research, development, and diffusion", In: Richard Schmalensee and Robert Willig, Editor(s), Handbook of Industrial Organization, Elsevier, Volume 1, Pages 849-908. (pdf)
  title={The timing of innovation: Research, development, and diffusion},
  author={Reinganum, Jennifer F},
  journal={Handbook of industrial organization},
  publisher={Amsterdam: North Holland}


The analysis of the timing of innovation posits a particular innovation (or sequence of innovations) and examines how the expected benefits, the cost of R&D and interactions among competing firms combine to determine the pattern of expenditure across firms and over time, the date of introduction, and the identity of the innovating firm. In the case of a sequence of innovations, the expected lifetime of a given innovation and the pattern of technological leadership are also determined endogenously. Given that an innovation has been perfected, the extent and timing of its dissemination into use may be examined. Again this may depend upon a number of factors, including the existence of rival firms and institutions which may facilitate or retard the dissemination of innovations...

Quick Summary

This paper is a chapter in the Handbook of Industrial Organization. It cover four basic topics:

  1. Symmetric Models - Firms are homogeneous and compete to produce an innovation and secure the associated rents
  2. Asymmetric Models - There is an incumbent and one or more entrants, which compete to produce an innovatation and secure the associated rents
  3. Licensing and Joint Ventures - Do joint ventures increase or restrict innovation and the dissemination of innovations? Licensing might be used strategically to disincentivize a rival from pursuing further innovation.
  4. Diffusion - What theories can help explain the observed patterns of diffusion for innovations? When will firms adopt too early or too late?

Symmetric Models

There are some basic assumptions that underlie all of the symmetric models:

  • Firms are homogeneous, and so we look for symmetric solutions
  • Innovation is costly, with costs (eventually at least) decreasing and convex (i.e. there are eventual diseconomies of scale)
  • The first to invent wins the rights to a rent (i.e. gets a patent) and everyone else gets nothing (i.e. no spillovers)


  • All of the models are in continuous time (though this need not be true)
  • The date of invention may be deterministic or stochastic (which results in perfectly or imperfectly discriminating contests)

The symmetric models section proceeds as follows:

  1. Simple first price or all pay auction set-ups are discussed
  2. Dasguspta and Stiglitz (1980) gives a first price auction like game, but with a continously discounted prize and bids translate into times to invention.
  3. Kamien and Schwatz (1976) describe the partial equilibrium of a patent race with stochastic invention
  4. Loury (1979) gives the full equilibrium result
  5. Lee and Wilde (1980) extend Loury by considering the payment of cost as continous until an invention occurs, rather than solely at the outset
  6. Reinganum (1982) gives an even more general model, which allows for variable rates of investment over time (and so full strategic responses to rival's investment), and also considers imperfect patent protection.