Stepwise Innovation by an Oligopoly

Article Source: International Journal of Industrial Organization, Vol. 61, pp. 413-438, 2018
Publication Date:
Time to Read: 2 minute read
Written By:

 Christian Riis

Christian Riis

 Erlend S. Riis

Erlend S. Riis



Antitrust authorities assess the effect of mergers on innovation. Sometimes, different economic models yield different predictions about the effects of competition on innovation.


Policy Relevance:

More sophisticated models improve assessment of the effect of mergers on innovation.


Key Takeaways:
  • Antitrust authorities assessing the effect of a proposed merger on innovation rates may use economic models of “stepwise” innovation; with stepwise innovation, laggard firms are assumed to be no more than one step behind the leading firms in innovation.
  • Technology spillovers reduce the costs of laggard firms in catching up with leaders; spillovers tend to moderate the effect of competition on innovation.
  • Philippe Aghion’s model of stepwise innovation simulates the effect of increased intensity of competition on innovation in a duopoly, but does not allow consideration of situations with more than two competing firms; duopoly models are too limited to be useful in real-world merger assessments.
  • A new “oligopoly” model simulates the effect of competition on innovation in a market with two or more rival firms.
    • As the number of firms increases, innovation at first increases, then levels off, then begins to fall.
    • If the number of rivals fall from three to two, innovation rates fall 16 to 20 percent.
    • Innovation rates peak in markets with three to five firms.
  • Depending on one’s assumptions, the oligopoly model yields different predictions than the duopoly model.
    • With no spillovers, the duopoly model predicts that more competition means less innovation, but the oligopoly model predicts that more rivals means more innovation.
    • With some spillovers, the duopoly model predicts that more competition means more innovation, but the oligopoly model predicts that innovation will at first increase as the number of rivals increases, but will level off and decrease as the number of rivals increases further.
  • The oligopoly model shows that a reduction in the number of rivals by merger can have an adverse effect on innovation even in an industry with five or six firms.
  • Real-world mergers involve factors not included in the oligopoly or duopoly models; mergers may create efficiencies such as the elimination of redundant research and development or benefits from complementary ventures.



Richard Gilbert

About Richard Gilbert

Richard Gilbert is a professor emeritus of economics at the University of California, Berkeley. He is also Chair of the Berkeley Competition Policy Center. Professor Gilbert's research specialties are in the areas of competition policy, intellectual property, and research and development.

Professor Gilbert was Chair of the Department of Economics at Berkeley from 2002 to 2005. Additionally, he worked as Deputy Assistant Attorney General for Economics in the U.S. Department of Justice’s Antitrust Division from 1993 to 1995.