The Economics of Socially-Efficient Privacy and Confidentiality Management for Statistical Agencies

Logo

John M ABOWD (Principal Investigator)
Ian M SCHMUTTE (Principal Investigator)
Lars VILHUBER (Principal Investigator)
*

This is the Final and Cumulative Annual Report for Alfred P. Sloan Foundation Grant G-2015-13903 ``The Economics of Socially-Efficient Privacy and Confidentiality Management for Statistical Agencies''.

Final report to Sloan - Grant G-2015-13903

This is the Final and Cumulative Annual Report for Alfred P. Sloan Foundation Grant G-2015-13903 ``The Economics of Socially-Efficient Privacy and Confidentiality Management for Statistical Agencies’’. The award was made in 2015, extended in 2018, and concluded in March 2019.

Original Project Goal

Overall, the proposal aimed to provide guidance primarily for public statistical agencies on the most promising methods to manage confidential information resources to maximal social benefit. We proposed to create a library of the most promising privacy-preserving algorithms, then systematically evaluate the trade-offs between accuracy and privacy when they are applied to real large-scale databases. Furthermore, we aimed to establish new measurements of the relative social preferences for privacy protection and data accuracy.

Expected Products

We set out to publish scientific papers as well as a toolkit of tested algorithms implemented in commonly used software. Our data products were to include real and simulated synthetic data. Potential collaborations may result in new or improved synthetic data products by the U.S. Census Bureau.

Outcomes

The grant enabled the publication of numerous articles, by ourselves and others. In particular, we published articles in various economics journals on … [summarize here]. We also released proceedings from three practitioner-oriented workshops on privacy, providing practical guidance on the implementation of privacy-oriented methods to various government agencies. Our research into algorithms was incorporated into the DPComp project, which has gone on to provide a great database of algorithms and datasets for privacy-oriented implementations. We also continued to support the Synthetic Data Server, and through that support, allow for academic papers to be published using synthetic data as their primary input, as well as allowing for the Census Bureau and other agencies to learn from the experience. Our ability to create new synthetic data products by the U.S. Census Bureau was vastly increased through Abowd’s appointment as Chief Scientist at the Census Bureau.

The full list of our activities and outputs can be found in the detailed “Final and Cumulative Annual Report for Alfred P. Sloan Foundation Grant G-2015-13903 `The Economics of Socially-Efficient Privacy and Confidentiality Management for Statistical Agencies’

We expected to be able to contribute to improving access to detailed but confidential databases, while preserving the privacy of respondents and subjects. The increase in access has a multiplier effect — future researchers will be able to make new findings, and better inform policy, and their results can be used to improve the published data.

While the impact of our work will take a few years to be felt, we have exceeded our own expectations in our ability to inform future researchers.