Abstract: Network effects generate additional utility for consumers based on user scale, which increases the importance of market share for platforms, intensifies competition among them, and may lead to suboptimal pricing. In contrast, consumers’ multi‑homing behavior—simultaneously purchasing multiple products—can mitigate such competitive pressure. Based on the Hotelling model, this paper constructs a multi‑platform game framework that incorporates consumers’ option to exit, product heterogeneity, and three types of power structures (cooperation, symmetric simultaneous pricing, and asymmetric sequential pricing), to systematically examine the impact of network effects and multi‑homing behavior on platforms’ pricing strategies and profits. Contrary to conventional findings that network effects undermine platforms’ prices and profits, the study shows that in incompletely covered markets where consumers can opt out, network effects do not affect pricing decisions but instead expand platform scale and enhance profits. After introducing multi‑homing behavior, network effects can, under certain conditions, promote platform profits, indicating that their impact mechanism depends critically on consumer participation patterns. A comparison with cooperative decisions reveals that network effects significantly amplify the profit loss in non‑cooperative games, thereby intensifying the prisoner’s dilemma in traditional pricing competition and highlighting the strategic value of cooperation among platforms in strong network‑effect environments. Furthermore, under asymmetric power structures, network effects weaken or even reverse the traditional second‑mover advantage, strengthening the market dominance of the first mover, while simultaneously exerting a significant negative impact on overall market efficiency.

Key words: network effects, multi-homing behavior, power structure, market efficiency