Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
How many US citizens have been accidentally deported by ICE since 2010?
Executive Summary
A Government Accountability Office (GAO) review found that at least 70 people ICE removed between 2015 and 2020 were potential U.S. citizens, while other analyses indicate the true figure could be higher because of poor record-keeping and identification errors. Broader datasets and investigative reporting document hundreds of arrests and detentions of people ICE and CBP flagged as U.S. citizens or potentially eligible for deportation across multiple years, leaving the total number of accidental deportations since 2010 unresolved [1] [2] [3].
1. Why the GAO count matters — and why it probably understates the scope
The GAO’s systematic review is the most authoritative, documented finding that ICE removed 70 potential U.S. citizens between 2015 and 2020, and that the agency arrested 674 and detained 121 in that window. The report also highlights structural flaws—inconsistent training, database errors, and no requirement to update citizenship fields after investigations—which produce undercounts in official tallies and make it impossible to produce a definitive number since 2010 from internal records alone [1] [2] [3]. That GAO timeline is narrow; it does not address earlier years. The report’s central finding is not merely the 70 removals but the systemic data and process failures that likely conceal additional wrongful actions.
2. Independent datasets that expand the picture beyond GAO
Longitudinal public records analyses and advocacy datasets capture far larger numbers of misidentifications and custody events. Transactional Records Access Clearinghouse (TRAC) analysis, cited alongside GAO findings, identified at least 2,840 people wrongly flagged as eligible for removal between 2002 and 2017, with at least 214 taken into custody, indicating that erroneous targeting has been recurring and widespread across multiple administrations [1] [3]. These figures do not map perfectly onto confirmed deportations, but they document extensive misclassification and detention that increase the plausibility of wrongful removals beyond the GAO’s 2015–2020 window.
3. Conflicting narratives from DHS and investigative reporters
Investigative reporting and watchdog work describe hundreds of detentions of U.S. citizens or people asserting citizenship, including ProPublica’s reporting that immigration agents held more than 170 Americans since the start of a recent administration, while DHS press statements have denied routine deportation of citizens and argued that some detentions were lawful arrests for obstruction or outstanding warrants [4] [5]. This divergence reflects different priorities: watchdogs emphasizing documented incidents and patterns, and DHS emphasizing legal constraints and case-specific defenses. Both narratives are supported by facts: documented detentions exist, and U.S. law restricts removal of citizens; the tension lies in classification, proof, and record transparency [4] [5].
4. Why answers for “since 2010” remain elusive
Multiple sources show incomplete, inconsistent recordkeeping across ICE and CBP and note that ICE systems and training allowed low-level decisions without adequate oversight, producing both arrests and removals of people later identified as citizens or likely citizens [1] [2]. The GAO’s focus on 2015–2020 leaves a gap for 2010–2014; TRAC and other analyses provide earlier data but use different methodologies and definitions, complicating aggregation [3]. Therefore, any single-number claim for 2010–present either understates uncertainty (if it cites only the 70 GAO removals) or risks double-counting or conflating detentions, arrests, and confirmed removals when combining datasets.
5. What the various counts actually represent — arrests, detentions, removals
Available figures represent different types of enforcement contact: ICE wrongly identified 2,840 people as removable (TRAC analysis) and GAO reported 674 arrests, 121 detentions, and 70 removals in 2015–2020. Investigative reporting documents additional citizen detentions and alleged deportations in other years [1] [3] [4]. These categories are not interchangeable: an arrest or detention does not equal a completed deportation, and misidentification numbers include people who were never removed. Any accurate assessment must separate these categories and note that the confirmed removals count remains the narrowest, best-documented subset.
6. Bottom line: a confirmed minimum and meaningful uncertainty above it
The most defensible, evidence-based statement is that ICE removed at least 70 individuals it later identified as potential U.S. citizens between 2015 and 2020, and that far larger numbers of arrests and detentions of people later flagged as citizens occurred across longer timeframes, suggesting the true total of accidental removals since 2010 is unknown and plausibly higher [1] [2] [3]. The discrepancy arises from systemic recordkeeping failures, divergent datasets, and differing definitions used in reporting and oversight, leaving the precise count since 2010 an open question supported by multiple, documented sources.