Working Paper: NBER ID: w30881
Authors: Lin William Cong; Simon Mayer
Abstract: In a model of data-driven firm competition, data are jointly produced by users, buying firms’ services and contributing data, and firms, investing in data infrastructure and collection. Data collection improves services, benefiting users, but may reduce competition, harming users. Dispersed users do not internalize the impact of their data contribution on (i) service quality, (ii) competition, and (iii) firms’ investment incentives, causing inefficient data over- or underinvestment. Unlike data sharing, user privacy protection policies, or data markets, a data union — which coordinates users’ data contributions — or data trust — which intermediates data sales — can address these inefficiencies.
Keywords: Data Union; User Privacy; Data Economy; Regulation
JEL Codes: L10; L41; L50; O30
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| User contributions to data collection (C81) | Service quality (L15) |
| User contributions to data collection (C81) | Competition (L13) |
| Competition (L13) | User welfare (I31) |
| Users do not internalize data contributions (D16) | Inefficient data over or underinvestment (G31) |
| Existing user privacy protection policies (D18) | User welfare (I31) |
| Data union (Y10) | User welfare (I31) |
| Data sharing policies (D16) | Competition (L13) |
| Data sharing policies (D16) | Data collection (C80) |
| Firms' investments in data infrastructure (D25) | Service improvement (O35) |
| Firms' investments in data infrastructure (D25) | Competition (L13) |