[Paper Review] Option contracts for a privacy-aware market

Maurizio Naldi, Giuseppe D’Acquisto|arXiv (Cornell University)|Sep 22, 2015

Auction Theory and Applications11 references3 citations

TL;DR

This paper proposes a privacy-aware marketplace where suppliers protect their private inventory data using differential privacy, while a broker uses option contracts to guarantee end customers full delivery. The broker profits by sourcing items from low-cost, privacy-protected suppliers and sets a minimum end price formula that accounts for uncertainty in available supply and privacy costs.

ABSTRACT

Suppliers (including companies and individual prosumers) may wish to protect their private information when selling items they have in stock. A market is envisaged where private information can be protected through the use of differential privacy and option contracts, while privacy-aware suppliers deliver their stock at a reduced price. In such a marketplace a broker acts as intermediary between privacy-aware suppliers and end customers, providing the extra items possibly needed to fully meet the customers' demand, while end customers book the items they need through an option contract. All stakeholders may benefit from such a marketplace. A formula is provided for the option price, and a budget equation is set for the mechanism to be profitable for the broker/producer.

Motivation & Objective

• To design a marketplace where suppliers can sell inventory without revealing private details like stock levels or identity.
• To address the challenge of uncertain supply due to differential privacy obfuscation in statistical databases.
• To enable a broker to profitably guarantee end-customer demand through option contracts and strategic procurement.
• To derive a pricing mechanism that ensures the broker remains profitable on average despite supply uncertainty.
• To balance privacy protection with economic incentives for suppliers, brokers, and end customers.

Proposed method

• Uses differential privacy via Laplace noise to obfuscate true supplier inventory levels, protecting supplier privacy.
• Employs option contracts where end customers pay a premium to secure delivery of k* items, regardless of actual supply.
• Derives an option price formula that accounts for excess inventory risk and privacy-induced uncertainty.
• Establishes a budget equation for the broker, incorporating query cost, supplier price, production cost, and option revenue.
• Derives a minimum end price formula that ensures profitability on average, using expected values and Laplace noise assumptions.
• Introduces a dependency of supplier price and query cost on the level of differential privacy, enabling trade-off optimization.

Experimental results

Research questions

• RQ1 How can a marketplace be structured to allow suppliers to sell inventory while preserving their privacy?
• RQ2 What is the optimal option price that protects the broker from financial risk due to uncertain supply from privacy-aware suppliers?
• RQ3 How can the broker ensure profitability when sourcing from privacy-protected suppliers at reduced prices?
• RQ4 What is the minimum end price that guarantees the broker's average profit, given supply uncertainty and privacy costs?
• RQ5 How does the level of differential privacy affect the pricing and economic incentives across all stakeholders?

Key findings

• The option price is derived as a function of the actual excess inventory and the noise level in the differential privacy mechanism, with the risk term vanishing as privacy decreases.
• The broker's budget equation ensures profitability by balancing query cost, supplier price, production cost, and option revenue.
• The minimum end price for customers is derived using expected values of supply and demand, incorporating Laplace noise distribution for uncertainty.
• The formula for the end price includes terms that account for the expected shortfall in supply from privacy-aware suppliers and the cost of producing the remainder.
• The model shows that suppliers can choose their privacy level (via noise variance) and trade it off against lower prices, enabling a privacy-profit trade-off.
• All stakeholders benefit: suppliers gain privacy and sales, brokers gain revenue from options and arbitrage, and customers get guaranteed delivery at lower prices.