Market Entry and Competition Under Network Effects

We consider a three-stage game in which, first, a large number of potential firms make entry decisions, then those who choose to stay in the market decide on the investment (quality) level in each product, and last, customers with heterogeneous preferences arrive sequentially to make (random) purchase decisions based on product quality and historical sales under the network effect according to a discrete choice model. We characterize such a random purchase process and show that a growing network effect always contributes to more sales concentration ex post on a small number of products. Perhaps surprisingly, we further show several phase-changing phenomena regarding equilibrium outcomes with respect to the network effect’s strength. In particular, the equilibrium product variety (respectively, quality investment) first decreases (respectively, increases) and then increases (respectively, decreases) as the network effect grows. Specifically, when the strength of the network effect is below a threshold, an increasing network effect would shift more sales toward those products with higher quality, preventing more products from entering the market ex ante and inducing firms to adopt the high-budget equilibrium strategy by making a small number of high-quality products, which is consistent with the blockbuster phenomenon. When the strength of the network effect is above the threshold, the network effect would easily cause the market to be concentrated on a few products ex post; even some low-quality products may have a chance to become a “hit.” Interestingly, in this case, when the network effect is growing, the ex ante equilibrium product variety will be wider, and firms adopt the low-budget equilibrium strategy by making a (relatively) large number of low-quality products, a finding consistent with the long tail theory. We then establish the robustness of the previous main insights by accounting for endogenized pricing and multiproducts carried by each firm.

Funding: Y. Feng was financially supported by the Major Program of National Natural Science Foundation of China [Grants 72192830 and 7219283X], Fundamental Research Funds for the Central Universities, and Program for Innovative Research of Shanghai University of Finance and Economics. M. Hu was supported by the Natural Sciences and Engineering Research Council of Canada [Grants RGPIN-2015-06757 and RGPIN-2021-04295].

Supplemental Material: The online appendix is available at https://doi.org/10.1287/opre.2022.0275.