Network Externality Products C & Competition Strategy: An Experimental Economics Approach

Geng Zhou, Jiaofei Zhong, Han Zhou, Yu ChenOpenalex Unknown 2007 · DOI · OpenAlex

Abstract Based on the theory of Complexity Adaptive System (CAS) and methods of experimental economics, using Multi-Agent-Based Simulation, we build a model of the network externality product market with the platform Netlogo3.1.2, analyze the macroscopic consumers’ behavior, find out the important aspects that influence the spreading of products and the proper strategies of enterprises during this process and make a prediction of the China’s instant messenger software market.

Introduction

In the modern digital economy, products like Microsoft Office or Tencent QQ often achieve market dominance that defies traditional feature-based competition. This phenomenon is driven by network externality, where a product's value increases exponentially with its user base—a concept formalized by Metcalfe’s Law ($V \propto n^2$). This paper explores the dynamic complexity of such markets using experimental economics and multi-agent-based simulation (MABS) to identify the strategic levers that lead to market monopoly.

Modeling Market Dynamics

To capture the "bottom-up" emergence of market trends, the researchers utilized Complexity Adaptive System (CAS) theory. Using the NetLogo 3.1.2 platform, they simulated a virtual marketplace with over 1,000 autonomous agents. Each agent operates based on specific rules of motion, influence (word-of-mouth and advertising), and purchasing thresholds.

Conceptual model of consumer behavior and market influence in a network externality environment.

The model tracks how "potential buyers" transition to "users" as their perceived value of a product reaches a set threshold, influenced by both external advertisements and the existing user network. This framework allows for the study of both single-product growth and competitive two-product scenarios where "lock-in" effects become prominent.

The Tipping Point and Information Flow

A central finding of the simulation is the existence of a "critical number" or tipping point. Growth is initially slow and linear; however, once the user base hits this threshold, the positive feedback loop of network effects triggers exponential growth.

Crucially, this "explosion" depends on the information circulation rate. The researchers found that if the rate at which agents communicate is too low, the network effect may never manifest, leaving the product in a state of stagnant, linear growth. High-speed communication mediums, such as the Internet, are therefore essential catalysts for products characterized by high network externalities.

Experimental Results

The simulations demonstrate that the strength of network externality directly dictates the speed of market capture. As shown in Figure 4, higher externality values significantly compress the time needed to reach the critical threshold and achieve full market occupation.

Impact of network externality on the time to reach critical mass and market saturation.

Furthermore, the research highlights the power of randomness in early-stage competition. In a two-product market with identical features, small random fluctuations in early adoption can allow one product to reach the critical mass first. This leads to a "winner-takes-all" outcome where the competitor is effectively locked out by the dominant player's established network value.

Strategic Implications

For enterprises, these findings suggest a shift in strategy depending on the market phase:

  1. Pre-Critical Mass: Focus on building an initial user base through aggressive discounts, free editions, and advertising basic functions to reach the tipping point faster.
  2. Post-Critical Mass: Shift investment toward "value-added services" and increasing "switching costs" (lock-in). At this stage, word-of-mouth and network value become more efficient drivers of growth than traditional advertising.

Conclusion and Future Outlook

The study validated its model against the historical growth of Tencent QQ, correctly predicting that while QQ would dominate the general IM market, specialized niches (such as enterprise-specific or web-integrated IMs) would remain open for competition. Future research could enhance these models by incorporating more complex social network topologies and real-world consumer data to move beyond qualitative abstractions.

Summary updated

Abstract

Based on the theory of Complexity Adaptive System (CAS) and methods of experimental economics, using Multi-Agent-Based Simulation, we build a model of the network externality product market with the platform Netlogo3.1.2, analyze the macroscopic consumers’ behavior, find out the important aspects that influence the spreading of products and the proper strategies of enterprises during this process and make a prediction of the China’s instant messenger software market.
BibTeX
@inproceedings{zhou2007network,
  author = {Geng Zhou and Jiaofei Zhong and Han Zhou and Yu Chen},
  title = {{Network Externality Products C \& Competition Strategy: An Experimental Economics Approach}},
  booktitle = {Openalex Unknown},
  year = {2007},
  doi = {10.1007/978-0-387-75466-6\_9},
}
Figure