{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T21:26:12Z","timestamp":1785360372154,"version":"3.55.0"},"reference-count":57,"publisher":"National Academy of Sciences","issue":"45","license":[{"start":{"date-parts":[[2025,11,5]],"date-time":"2025-11-05T00:00:00Z","timestamp":1762300800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":["www.pnas.org"],"crossmark-restriction":true},"short-container-title":["Proc. Natl. Acad. Sci. U.S.A."],"published-print":{"date-parts":[[2025,11,11]]},"abstract":"<jats:p>\n                    The US Decennial Census serves as the foundation for many high-profile policy decision-making processes, including federal funding allocation and redistricting. In 2020, the Census Bureau adopted differential privacy to protect the confidentiality of individual responses through a disclosure avoidance system that injects noise into census data tabulations. The Bureau subsequently posed an open question: Could stronger privacy guarantees be obtained for the 2020 US Census compared to their published guarantees, or equivalently, had the privacy budgets been fully utilized? In this paper, we address this question affirmatively by demonstrating that the 2020 US Census provides significantly stronger privacy protections than the officially published guarantees suggest at each of the eight geographical levels, from the national level down to the block level. This finding is enabled by our precise tracking of privacy losses using\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" display=\"inline\" overflow=\"scroll\">\n                        <mml:mi>f<\/mml:mi>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    -differential privacy, applied to the composition of private queries across these geographical levels. Our analysis reveals that the Census Bureau introduced unnecessarily high levels of noise to meet the specified privacy guarantees for the 2020 Census. Consequently, we show that noise variances could be reduced by 15.08% to 24.82% while maintaining nearly the same level of privacy protection for each geographical level, thereby improving the accuracy of privatized census statistics. We empirically demonstrate that reducing noise injection into census statistics mitigates distortion caused by privacy constraints in downstream applications of private census data, illustrated through a study examining the relationship between earnings and education.\n                  <\/jats:p>","DOI":"10.1073\/pnas.2500337122","type":"journal-article","created":{"date-parts":[[2025,11,5]],"date-time":"2025-11-05T17:47:06Z","timestamp":1762364826000},"update-policy":"https:\/\/doi.org\/10.1073\/pnas.cm10313","source":"Crossref","is-referenced-by-count":2,"title":["The 2020 US Decennial Census is more private than you (might) think"],"prefix":"10.1073","volume":"122","author":[{"given":"Buxin","family":"Su","sequence":"first","affiliation":[{"name":"Applied Mathematics and Computational Science","place":["Philadelphia"]},{"name":"University of Pennsylvania","place":["Philadelphia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1787-1219","authenticated-orcid":false,"given":"Weijie J.","family":"Su","sequence":"additional","affiliation":[{"name":"Wharton Statistics and Data Science Department","place":["Philadelphia"]},{"name":"University of Pennsylvania","place":["Philadelphia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chendi","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Statistics and Data Science","place":["Xiamen, Fujian, China"]},{"name":"School of Economics","place":["Xiamen, Fujian, China"]},{"name":"Xiamen University","place":["Xiamen, Fujian, China"]},{"name":"Ministry of Education lab of Key Laboratory of Econometrics and Wang Yanan Institute for Studies in Economics","place":["Xiamen, Fujian, China"]},{"name":"Xiamen University","place":["Xiamen, Fujian, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"341","published-online":{"date-parts":[[2025,11,5]]},"reference":[{"key":"e_1_3_4_1_2","volume-title":"Uses of Census Bureau Data in Federal Funds Distribution: A New Design for the 21st Century","author":"Hotchkiss M.","year":"2017","unstructured":"M. Hotchkiss, J. Phelan, Uses of Census Bureau Data in Federal Funds Distribution: A New Design for the 21st Century (United States Census Bureau, 2017)."},{"key":"e_1_3_4_2_2","unstructured":"US Census Bureau Census bureau data guide more than 2.8 trillion in federal funding in fiscal year 2021 (2023). https:\/\/www.census.gov\/newsroom\/press-releases\/2023\/decennial-census-federal-funds-distribution.html. Accessed 1 October 2024."},{"key":"e_1_3_4_3_2","volume-title":"Comment: The Essential Role of Policy Evaluation for the 2020 Census DisclosureAvoidance System","author":"Kenny C. T.","year":"2023","unstructured":"C. T. Kenny , Comment: The Essential Role of Policy Evaluation for the 2020 Census DisclosureAvoidance System (Harvard Data Science Review, 2023)."},{"key":"e_1_3_4_4_2","unstructured":"A. Cohen M. Duchin J. Matthews B. Suwal \u201cCensus TopDown: The impacts of differential privacy on redistricting\u201d in 2nd Symposium on Foundations of Responsible Computing FORC 2021 June 9-11 2021 Virtual Conference LIPIcs K. Ligett S. Gupta Eds. (Schloss Dagstuhl - Leibniz-Zentrum f\u00fcr Informatik 2021) vol. 192 pp. 5:1\u20135:22."},{"key":"e_1_3_4_5_2","doi-asserted-by":"publisher","DOI":"10.1162\/00335530360535171"},{"key":"e_1_3_4_6_2","unstructured":"US Census Bureau Guidance for labor force statistics data users (2021). https:\/\/www.census.gov\/topics\/employment\/labor-force\/guidance.html. Accessed 1 October 2024."},{"key":"e_1_3_4_7_2","unstructured":"S. J. Eckman Apportionment and redistricting following the 2020 census (2021). https:\/\/sgp.fas.org\/crs\/misc\/IN11360.pdf. Accessed 1 October 2024."},{"key":"e_1_3_4_8_2","unstructured":"US Census Bureau 2020 Census apportionment results (2021). https:\/\/www.census.gov\/data\/tables\/2020\/dec\/2020-apportionment-data.html. Accessed 1 October 2024."},{"key":"e_1_3_4_9_2","doi-asserted-by":"publisher","DOI":"10.1080\/07350015.1989.10509729"},{"key":"e_1_3_4_10_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2218605120"},{"key":"e_1_3_4_11_2","unstructured":"M. Hawes Reconstruction and re-identification of the demographic and housing characteristics file (dhc) (2022). https:\/\/www2.census.gov\/about\/partners\/cac\/sac\/meetings\/2022-09\/presentation-reconstruction-and-re-dentification-of-dhc-file.pdf. Accessed 5 May 2024."},{"key":"e_1_3_4_12_2","unstructured":"J. M. Abowd Staring down the database reconstruction theorem (2019). https:\/\/www.census.gov\/content\/dam\/Census\/newsroom\/press-kits\/2018\/jsm\/jsm-presentation-database-reconstruction.pdf. Accessed 5 May 2024."},{"key":"e_1_3_4_13_2","unstructured":"J. Abowd Declaration of John Abowd in case no. 3: 21-cv-00211-rah-ecm-kcn the state of Alabama v. United States Department of Commerce (2021). https:\/\/www2.census.gov\/about\/policies\/foia\/records\/alabama-vs-doc\/alabama-ii-41-defs-pi-opposition-and-declarations.pdf. Accessed 2 October 2024."},{"key":"e_1_3_4_14_2","first-page":"265","article-title":"Calibrating noise to sensitivity in private data analysis","volume":"3876","author":"Dwork C.","year":"2006","unstructured":"C. Dwork, F. McSherry, K. Nissim, A. Smith, Calibrating noise to sensitivity in private data analysis. Theory Cryptogr. Proc. 3876, 265\u2013284 (2006).","journal-title":"Theory Cryptogr. Proc."},{"key":"e_1_3_4_15_2","doi-asserted-by":"crossref","unstructured":"C. Dwork K. Kenthapadi F. McSherry I. Mironov M. Naor \u201cOur data ourselves: Privacy via distributed noise generation\u201d in Advances in Cryptology-EUROCRYPT 2006: 24th Annual International Conference on the Theory and Applications of Cryptographic Techniques St. Petersburg Russia May 28-June 1 2006. Proceedings 25 S. Vaudenay Ed. (Springer 2006) pp. 486\u2013503.","DOI":"10.1007\/11761679_29"},{"key":"e_1_3_4_16_2","doi-asserted-by":"publisher","DOI":"10.1162\/99608f92.529e3cb9"},{"key":"e_1_3_4_17_2","unstructured":"J. M. Abowd et al. \u201cInvited lecture: The U.S. census bureau adopts differential privacy\u201d in KDD \u201918: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining Y Guo F. Farooq Eds. (ACM 2022)."},{"key":"e_1_3_4_18_2","unstructured":"V. Phillips Phillips v. U.S. Census Bureau (2023). https:\/\/thearp.org\/litigation\/phillips-v-us-census-bureau\/. Accessed 4 October 2024."},{"key":"e_1_3_4_19_2","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.abk3283"},{"key":"e_1_3_4_20_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2321196121"},{"key":"e_1_3_4_21_2","doi-asserted-by":"publisher","DOI":"10.2307\/j.ctvb1htjr"},{"key":"e_1_3_4_22_2","volume-title":"Differential Perspectives: Epistemic Disconnects Surrounding the U.S. Census Bureau\u2019s Use of Differential Privacy","author":"Boyd D.","year":"2022","unstructured":"D. Boyd, J. Sarathy, Differential Perspectives: Epistemic Disconnects Surrounding the U.S. Census Bureau\u2019s Use of Differential Privacy (Harvard Data Science Review, 2022)."},{"key":"e_1_3_4_23_2","unstructured":"D. Kifer et al. Bayesian and frequentist semantics for common variations of differential privacy. arXiv [Preprint] (2022). https:\/\/doi.org\/10.48550\/arXiv.2209.03310 (Accessed 4 October 2024)."},{"key":"e_1_3_4_24_2","doi-asserted-by":"publisher","DOI":"10.1111\/rssb.12454"},{"key":"e_1_3_4_25_2","first-page":"e726","article-title":"Privacy profiles and amplification by subsampling","volume":"10","author":"Balle B.","year":"2020","unstructured":"B. Balle, G. Barthe, M. Gaboardi, Privacy profiles and amplification by subsampling. J. Priv. Confidentiality 10, e726 (2020).","journal-title":"J. Priv. Confidentiality"},{"key":"e_1_3_4_26_2","unstructured":"US Census Bureau Privacy-protected 2010 census demonstration data | IPUMS NHGIS (2022). https:\/\/www.nhgis.org\/privacy-protected-2010-census-demonstration-data#v20220825-files. Accessed 2 January 2024."},{"key":"e_1_3_4_27_2","unstructured":"US Census Bureau Educational Attainment (2020). https:\/\/data.census.gov\/table\/ACSST5Y2020.S1501. Accessed 25 May 2024."},{"key":"e_1_3_4_28_2","unstructured":"C. L. Canonne G. Kamath T. Steinke \u201cThe discrete Gaussian for differential privacy\u201d in Advances in Neural Information Processing Systems H. Larochelle M. Ranzato R. Hadsell M. F. Balcan H. Lin Eds. (Curran Associates Inc. 2020) vol. 33 pp. 15676\u201315688."},{"key":"e_1_3_4_29_2","unstructured":"C. Dwork G. N. Rothblum Concentrated differential privacy. arXiv [Preprint] (2016). http:\/\/arxiv.org\/abs\/1603.01887 (Accessed 4 October 2024)."},{"key":"e_1_3_4_30_2","doi-asserted-by":"crossref","unstructured":"M. Bun T. Steinke \u201cConcentrated differential privacy: Simplifications extensions and lower bounds\u201d in Theory of Cryptography Conference M. Hirt A. D. Smith Eds. (Springer 2016) pp. 635\u2013658.","DOI":"10.1007\/978-3-662-53641-4_24"},{"key":"e_1_3_4_31_2","doi-asserted-by":"crossref","unstructured":"M. Bun C. Dwork G. N. Rothblum T. Steinke \u201cComposable and versatile privacy via truncated CDP\u201d in Proceedings of the 50th Annual ACM SIGACT Symposium on Theory of Computing I. Diakonikolas D. Kempe M. Henzinger Eds. (ACM 2018) pp. 74\u201386.","DOI":"10.1145\/3188745.3188946"},{"key":"e_1_3_4_32_2","doi-asserted-by":"publisher","DOI":"10.1137\/S0097539705447360"},{"key":"e_1_3_4_33_2","unstructured":"US Census Bureau Privacy-loss budget allocation 2022-08-25 (2022). https:\/\/www2.census.gov\/programs-surveys\/decennial\/2020\/program-management\/data-product-planning\/2010-demonstration-data-products\/02-Demographic_and_Housing_Characteristics\/2022-08-25_Summary_File\/2022-08-25_Privacy-Loss_Budget_Allocations.pdf. Accessed 2 January 2024."},{"key":"e_1_3_4_34_2","doi-asserted-by":"publisher","DOI":"10.1145\/1810891.1810916"},{"key":"e_1_3_4_35_2","doi-asserted-by":"publisher","DOI":"10.2478\/popets-2022-0013"},{"key":"e_1_3_4_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2017.2685505"},{"key":"e_1_3_4_37_2","doi-asserted-by":"crossref","unstructured":"I. Mironov \u201cR\u00e9nyi differential privacy\u201d in 2017 IEEE 30th Computer Security Foundations Symposium (CSF) B. K\u00f6pf S. Chong Eds. (IEEE 2017) pp. 263\u2013275.","DOI":"10.1109\/CSF.2017.11"},{"key":"e_1_3_4_38_2","first-page":"10","article-title":"Deep learning with Gaussian differential privacy","volume":"2020","author":"Bu Z.","year":"2020","unstructured":"Z. Bu, J. Dong, Q. Long, W. J. Su, Deep learning with Gaussian differential privacy. Harv. Data Sci. Rev. 2020, 10\u20131162 (2020).","journal-title":"Harv. Data Sci. Rev."},{"key":"e_1_3_4_39_2","first-page":"55051","article-title":"Unified enhancement of privacy bounds for mixture mechanisms via f-differential privacy","volume":"36","author":"Wang C.","year":"2024","unstructured":"C. Wang, B. Su, J. Ye, R. Shokri, W. J. Su, Unified enhancement of privacy bounds for mixture mechanisms via f-differential privacy. Adv. Neural Inf. Process. Syst. 36, 55051\u201355063 (2024).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"e_1_3_4_40_2","unstructured":"W. J. Su A statistical viewpoint on differential privacy: Hypothesis testing representation and Blackwell\u2019s theorem. arXiv [Preprint] (2024). http:\/\/arxiv.org\/abs\/2409.09558 (Accessed 4 October 2024)."},{"key":"e_1_3_4_41_2","unstructured":"US Census Bureau Selected Social Characteristics in the United States (U.S. Census Bureau 2020). Accessed 4 October 2024."},{"key":"e_1_3_4_42_2","unstructured":"US Census Bureau Selected Economic Characteristics (U.S. Census Bureau 2020). Accessed 4 October 2024."},{"key":"e_1_3_4_43_2","unstructured":"US Census Bureau Selected Housing Characteristics (U.S. Census Bureau 2020). Accessed 4 October 2024."},{"key":"e_1_3_4_44_2","unstructured":"US Census Bureau ACS Demographic and Housing Estimates (U.S. Census Bureau 2020). Accessed 4 October 2024."},{"key":"e_1_3_4_45_2","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.324.7328.23"},{"key":"e_1_3_4_46_2","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.adl2524"},{"key":"e_1_3_4_47_2","unstructured":"R. Cumings-Menon Full-information estimation for hierarchical data. arXiv [Preprint] (2024). http:\/\/arxiv.org\/abs\/2404.13164 (Accessed 4 October 2024)."},{"key":"e_1_3_4_48_2","doi-asserted-by":"publisher","DOI":"10.1093\/jssam\/smac021"},{"key":"e_1_3_4_49_2","unstructured":"J. Awan A. Edwards P. Bartholomew A. Sillers Best linear unbiased estimate from privatized histograms. arXiv [Preprint] (2024). http:\/\/arxiv.org\/abs\/2409.04387 (Accessed 4 October 2024)."},{"key":"e_1_3_4_50_2","first-page":"c871f9e0","article-title":"Coming to our census: How social statistics underpin our democracy (and republic)","volume":"2","author":"Sullivan T. A.","year":"2020","unstructured":"T. A. Sullivan, Coming to our census: How social statistics underpin our democracy (and republic). Harv. Data Sci. Rev. 2, c871f9e0 (2020).","journal-title":"Harv. Data Sci. Rev."},{"key":"e_1_3_4_51_2","volume-title":"The End of Inequality: One Person, One Vote and the Transformation of American Politics, Issues in American democracy","author":"Ansolabehere S.","year":"2008","unstructured":"S. Ansolabehere, J. Snyder, The End of Inequality: One Person, One Vote and the Transformation of American Politics, Issues in American democracy (Norton, 2008)."},{"key":"e_1_3_4_52_2","first-page":"e875","article-title":"Geographic spines in the 2020 census disclosure avoidance system","volume":"14","author":"Cumings-Menon R.","year":"2024","unstructured":"R. Cumings-Menon , Geographic spines in the 2020 census disclosure avoidance system. J. Priv. Confidentiality 14, e875 (2024).","journal-title":"J. Priv. Confidentiality"},{"key":"e_1_3_4_53_2","volume-title":"DAS-Implementation-Details","year":"2023","unstructured":"US Census Bureau, DAS-Implementation-Details (U.S. Census Bureau, 2023)."},{"key":"e_1_3_4_54_2","unstructured":"P. Kairouz Z. Liu T. Steinke \u201cThe distributed discrete gaussian mechanism for federated learning with secure aggregation\u201d in Proceedings of the 38th International Conference on Machine Learning ICML 2021 18-24 July 2021 Virtual Event Proceedings of Machine Learning Research M. Meila T. Zhang Eds. (PMLR 2021) vol. 139 pp. 5201\u20135212."},{"key":"e_1_3_4_55_2","unstructured":"Y. Zhu J. Dong Y. X. Wang \u201cOptimal accounting of differential privacy via characteristic function\u201d in International Conference on Artificial Intelligence and Statistics G. Camps-Valls F. J. R. Ruiz I. Valera Eds. (PMLR 2022) pp. 4782\u20134817."},{"key":"e_1_3_4_56_2","unstructured":"A. Koskela J. J\u00e4lk\u00f6 A. Honkela \u201cComputing tight differential privacy guarantees using FFT\u201d in The 23rd International Conference on Artificial Intelligence and Statistics AISTATS 2020 26-28 August 2020 Online [Palermo Sicily Italy] Proceedings of Machine Learning Research S. Chiappa R. Calandra Eds. (PMLR 2020) vol. 108 pp. 2560\u20132569."},{"key":"e_1_3_4_57_2","unstructured":"S. Gopi Y. T. Lee L. Wutschitz \u201cNumerical composition of differential privacy\u201d in Advances in Neural Information Processing Systems 34: Annual Conference on Neural Information Processing Systems 2021 NeurIPS 2021 December 6-14 2021 Virtual M. Ranzato A. Beygelzimer Y. N. Dauphin P. Liang J. W. Vaughan Eds. (Curran Associates Inc. 2021) pp. 11631\u201311642."}],"container-title":["Proceedings of the National Academy of Sciences"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pnas.org\/doi\/pdf\/10.1073\/pnas.2500337122","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,5]],"date-time":"2025-11-05T17:47:24Z","timestamp":1762364844000},"score":1,"resource":{"primary":{"URL":"https:\/\/pnas.org\/doi\/10.1073\/pnas.2500337122"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,5]]},"references-count":57,"journal-issue":{"issue":"45","published-print":{"date-parts":[[2025,11,11]]}},"alternative-id":["10.1073\/pnas.2500337122"],"URL":"https:\/\/doi.org\/10.1073\/pnas.2500337122","relation":{},"ISSN":["0027-8424","1091-6490"],"issn-type":[{"value":"0027-8424","type":"print"},{"value":"1091-6490","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,5]]},"assertion":[{"value":"2025-01-06","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-09-15","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-11-05","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e2500337122"}}