Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Data Trade and Consumer Privacy

Jiadong Gu

TL;DR

This paper develops an integrated mechanism-design model of data collection and trading across three markets: service (data sourcing), data (trade), and product (pricing). It shows that the informativeness of data trade is endogenously determined by the relative horizontal differentiation in the service market and the vertical differentiation in the product market, yielding regimes with no data trade, partial data trade, or full information disclosure. A key finding is that reducing the intermediary's bargaining power can curb information disclosure without harming welfare, while banning data trade can reduce social welfare when data enables downstream price discrimination. The framework thus implies that market-power regulation may be more effective than data-sharing regulation for protecting privacy and maintaining welfare, and it provides a precise characterization of when data sharing improves efficiency and when it primarily redistributes surplus. These results have practical relevance for policy debates on privacy regulation and platform governance in data-driven economies.

Abstract

This paper studies optimal mechanisms for collecting and trading data. Consumers benefit from revealing information about their tastes to a service provider because this improves the service. However, the information is also valuable to a third party as it may extract more revenue from the consumer in another market called the product market. The paper characterizes the constrained optimal mechanism for the service provider subject to incentive feasibility. It is shown that although the service provider sometimes sells no information or only partial information in order to preserve profits in the service market, selling full information is optimal when the data-sourcing market is highly differentiated. Moreover, a ban on data trade may reduce social welfare because it makes it harder to price discriminate in the product market. Instead, reducing the intermediary's bargaining power can protect privacy without hurting social welfare, which suggests that the regulation of market power is more efficient than the regulation of data sharing.

Data Trade and Consumer Privacy

TL;DR

This paper develops an integrated mechanism-design model of data collection and trading across three markets: service (data sourcing), data (trade), and product (pricing). It shows that the informativeness of data trade is endogenously determined by the relative horizontal differentiation in the service market and the vertical differentiation in the product market, yielding regimes with no data trade, partial data trade, or full information disclosure. A key finding is that reducing the intermediary's bargaining power can curb information disclosure without harming welfare, while banning data trade can reduce social welfare when data enables downstream price discrimination. The framework thus implies that market-power regulation may be more effective than data-sharing regulation for protecting privacy and maintaining welfare, and it provides a precise characterization of when data sharing improves efficiency and when it primarily redistributes surplus. These results have practical relevance for policy debates on privacy regulation and platform governance in data-driven economies.

Abstract

This paper studies optimal mechanisms for collecting and trading data. Consumers benefit from revealing information about their tastes to a service provider because this improves the service. However, the information is also valuable to a third party as it may extract more revenue from the consumer in another market called the product market. The paper characterizes the constrained optimal mechanism for the service provider subject to incentive feasibility. It is shown that although the service provider sometimes sells no information or only partial information in order to preserve profits in the service market, selling full information is optimal when the data-sourcing market is highly differentiated. Moreover, a ban on data trade may reduce social welfare because it makes it harder to price discriminate in the product market. Instead, reducing the intermediary's bargaining power can protect privacy without hurting social welfare, which suggests that the regulation of market power is more efficient than the regulation of data sharing.
Paper Structure (62 sections, 19 theorems, 154 equations, 3 figures, 2 tables)

This paper contains 62 sections, 19 theorems, 154 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Literature.
  3. Outline.
  4. The Model
  5. Timing.
  6. Discussion about Modeling Assumption.
  7. Preliminaries
  8. Interim Payoff of Producer.
  9. Interim Payoff of Consumers.
  10. Producer Optimality
  11. Consumers Optimality
  12. Consumers of Type $H$.
  13. Consumers of Type $L$.
  14. Intermediary Optimality
  15. Optimal Design
  16. ...and 47 more sections

Key Result

Lemma 1

The trade data consists of two signals $\mathcal{S} =\left\{l,h\right\}$ with

Figures (3)

  • Figure 1: The overview of the modeling ingredients and interactions among players.
  • Figure 2: Timing of the events.
  • Figure 3: The triangle gray part is the feasible data trade gain from a data trade. Given $\mu_0 > v_L/v_H$: the product revenue is $\mu_0v_H$ at point $A$ without data trade. It goes up to point $B$ under a partial information data which induces posteriors $\mu_l$ and $\mu_h$. If the data is of full information, the product revenue is $\mu_0 v_H + \left( 1- \mu_0 \right) v_L$.

Theorems & Definitions (34)

  • Lemma 1: Direct Data Trade
  • Lemma 2
  • Proposition 1: Optimal data— binary service
  • Proposition 2: Optimal design— small $\mu_0$
  • Proposition 3: Optimal design— large $\mu_0$
  • Corollary 1: Full privacy
  • Corollary 2: Full disclosure
  • Lemma 3: Efficient mechanism
  • Lemma 4: Results without data market
  • Proposition 4: Regulation of data sharing
  • ...and 24 more