Date of Submission


Date of Award


Institute Name (Publisher)

Indian Statistical Institute

Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy

Subject Name

Quantitative Economics


Economic Research Unit (ERU-Kolkata)


Mitra, Manipushpak (ERU-Kolkata; ISI)

Abstract (Summary of the Work)

Industrial economists are often interested in comparing different market structures which are primarily based on their market outcomes and then try to determine the best market structure considering either the society’s welfare or the firm’s profit and sometimes considering both. In this context, the "Cournot-Bertrand comparison" is one such important comparison that has often been analyzed in the literature of industrial economics. The main structural difference between Cournot competition and Bertrand competition arises due to the strategic variable through which firms interact with each other in the market. To be more specific, in case of Cournot competition, firms compete with quantities while under the Bertrand competition they compete with prices. The first study with differentiated products was made by Singh and Vives (1984). They conclude that under Cournot duopoly each firm in the industry produces less, charges more and earns higher profit than under Bertrand duopoly. Further, they argued that the latter is efficient than the former in terms of welfare ranking. We refer to these rankings as the standard rankings. Subsequent studies in this literature have mainly concentrated in determining the circumstances where these standard rankings are either partially reversed or fully reversed. One such contribution by Häckner (2000) shows that the standard rankings are dependent on the duopoly assumption and they get reversed under sufficient quality differences with increasing number of firms. How ever they do not consider the welfare rankings between Cournot and Bertrand. Hsu and Wang (2005) conclude that the standard rankings hold in case of welfare with any number of firms. Amir and Jin (2001), have extended the "Cournot-Bertrand comparison" by including the following market indicators:- mark-up output ratio, average output, average price and Herfindahl index. Except for Singh and Vives (1984), the aforementioned studies deals with oligopoly market with linear demand. On the other hand, Vives (1985) and Okuguchi (1987) have worked with oligopoly markets assuming general non-linear demand functions. Subsequent studies by Mukherjee (2005) and Cellini et al. (2004) for free entry; Symeonidis (2003) and Lin and Saggi (2002) for endogenous Research & Development expenditure; López and Naylor (2004) for the wage bargaining provided evidence on partial reversal of the standard rankings. Arya et al. (2008b) and Alipranti et al. (2014) have shown the complete reversal of the standard rankings with a vertically related producer along with Ghosh and Mitra (2009) who get the same with mixed market. One important contribution with homogeneous product is by Dastidar (1997) where it is established that the standard Bertrand-Cournot rankings are sensitive to the market sharing rules.2So far the discussion has been primarily based on separate analysis of both Cournot and Bertrand competition. However, this separate analysis is rigid in the sense that the strategic variable through which firms compete in the market is exogenously given. But it may also be possible that firms can endogenously determine their strategic variables. Consequently, the equilibrium outcome may be one of the following: (i) only Cournot competition (ii) only Bertrand competition (iii) neither Cournot nor Bertrand competition.


ProQuest Collection ID:

Control Number


Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.


Included in

Mathematics Commons