The Impact of Multi-Homing on Market Concentration and Consumer Welfare in Two-Sided Digital Platforms: An Empirical Analysis of the Chinese E-commerce Sector
Description
This study on multi-homing in Chinese e-commerce uses diverse datasets to analyze market dynamics, consumer behavior, and policy impacts. The data, collected from Q1 2019 to Q4 2023, includes: Consumer records: A sample of 50 out of 10,000 anonymized consumer profiles for Q4 2023, gathered via public APIs and surveys. It covers platform usage, purchases, and satisfaction ratings, crucial for understanding multi-homing patterns and market effects. Seller data: 50 out of 5,000 anonymized seller records for Q4 2023, obtained through surveys and marketplace analytics. It includes platform activity, sales, revenue, and listings, vital for analyzing seller strategies and market competition. Price data: 50 out of 100,000 quarterly product records, collected via web scraping and APIs. It covers prices, categories, and sales volumes across platforms, essential for examining pricing strategies and product diversity. Platform revenue: Quarterly data on GMV, commission, and advertising revenue for major platforms, compiled from public reports. This non-confidential information is key to assessing market size, growth trends, and platform performance. Consumer surveys: 50 out of 10,000 quarterly responses on platform preferences and demographics. This anonymous data helps correlate factors influencing multi-homing behavior across consumer segments. Policy simulations: 50 out of 1000 scenarios based on economic models and empirical data, crucial for predicting policy impacts on multi-homing and market outcomes. Moderation analysis: Quarterly data on multi-homing rates, network effects, market concentration, prices, and product variety. Derived from platform statistics and research reports, it's vital for examining market factor interactions. Platform-level PVAR data: Quarterly information on market share, prices, product variety, user satisfaction, and activity. Compiled from public reports and aggregated user data, it's essential for analyzing dynamic relationships between multi-homing, market structure, and consumer welfare. All datasets are non-confidential and anonymized where necessary. They combine to provide a view of the Chinese e-commerce ecosystem, enabling in-depth analysis of multi-homing effects on market concentration and consumer welfare. The data was collected through various methods including public APIs, web scraping, surveys, and analysis of public financial reports. It represents a mix of primary research conducted by the authors and aggregation of publicly available information.