An Economic Analysis of Privacy Protection and Statistical Accuracy as Social Choices

Abstract

Statistical agencies face a dual mandate to publish accurate statistics while protecting respondent privacy. Increasing privacy protection requires decreased accuracy. Recognizing this as a resource allocation problem, we propose an economic solution: operate where the marginal cost of increasing privacy equals the marginal benefit. Our model of production, from computer science, assumes data are published using an efficient differentially private algorithm. Optimal choice weighs the demand for accurate statistics against the demand for privacy. Examples from U.S.\ statistical programs show how our framework can guide decision-making. Further progress requires a better understanding of willingness-to-pay for privacy and statistical accuracy.

Forthcoming in American Economic Review

References in corpus (4)

• Scalable Privacy-Preserving Data Sharing Methodology for Genome-Wide Association Studies
• On the `Semantics' of Differential Privacy: A Bayesian Formulation
• The Geometry of Differential Privacy: the Sparse and Approximate Cases
• Optimizing error of high-dimensional statistical queries under differential privacy

Cited by in corpus (14)

• Privacy Protection, Measurement Error, and the Integration of Remote Sensing and Socioeconomic Survey Data
• Guidelines for Implementing and Auditing Differentially Private Systems
• Confidentiality Protection in the 2020 US Census of Population and Housing
• Differential Privacy for Government Agencies -- Are We There Yet?
• Assessing differentially private deep learning with Membership Inference
• Reproducibility and Transparency versus Privacy and Confidentiality: Reflections from a Data Editor
• Fair Decision Making using Privacy-Protected Data
• Decision Making with Differential Privacy under a Fairness Lens
• General Inferential Limits Under Differential and Pufferfish Privacy
• Big data, differential privacy, and national statistical organisations
• Why Data Anonymization Has Not Taken Off
• On Heuristic Models, Assumptions, and Parameters
• Suboptimal Provision of Privacy and Statistical Accuracy When They are Public Goods
• Protecting Time Series Data with Minimal Forecast Loss