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.
Fact check: What percentage of people that ICE are detaining are illegal immigrants?
Executive Summary
Recent government and independent analyses of ICE detention in late 2024–mid 2025 show a substantial share of people held by ICE have no criminal record, with multiple datasets indicating roughly about a quarter to a third of those in detention lack criminal convictions—while some analyses report even higher shares when counting arrests over a broader period. These figures challenge the claim that ICE detains principally “criminal aliens,” but interpretation depends on which denominator and time window are used [1] [2] [3].
1. Data Point: The headline numbers that changed the debate
Government tallies published in September 2025 report that among roughly 59,762 people in ICE custody, 16,523 had no criminal record, 15,725 had a criminal record, and 13,767 had pending charges, which implies about 27.6% of detainees had no criminal convictions at that snapshot in time. This shift made non‑criminal detainees the single largest identifiable subgroup in custody and was highlighted as a factual counterpoint to policies framed around prioritizing criminal aliens [1]. The numbers are precise to that reporting period and reflect ICE’s internal categorizations.
2. Contrasting calculations: arrests vs. people in custody
Independent analyses using broader arrest windows find different shares. One report states that nearly a third of people arrested and booked into ICE detention had no criminal history, citing 11,700 people in detention without prior charges or convictions in a particular dataset, and noting large percentage increases relative to earlier years. Whether the share is ~28% or closer to one‑third depends on whether analysts use a point‑in‑time custody roster or aggregated arrests over months, and on how “no criminal record” is defined [2].
3. Wider context: longitudinal studies show larger shifts
A Cato Institute analysis cited in the assembled material examines October 2024–June 2025 arrests and finds about 65% of arrested individuals (133,000 people) had no criminal record, a markedly higher figure than the point‑in‑time custody snapshot. This suggests that when expanded to cover many arrests over time, the proportion of non‑criminals subject to ICE action can be substantially higher, reflecting operational focus, case outcomes, and churn in the detention system. That larger figure highlights how different methodological choices produce divergent headline claims [3].
4. What definitions and denominators change the story
Discrepancies stem from three methodological choices: (a) whether the analysis uses a point‑in‑time detention count versus aggregated arrests/bookings, (b) how “criminal record” is operationalized (convictions only, pending charges included, or prior arrests), and (c) the time period examined. Reports that use a larger arrest window or include people without convictions but with prior criminal arrests will yield different percentages. The assembled materials explicitly note these definitional differences and their impact on conclusions [1] [3].
5. Political framing: competing narratives and evident agendas
Media and advocacy narratives diverge: some articles frame the data as evidence that ICE is detaining large numbers of non‑criminal migrants, challenging rhetoric about targeting “the worst of the worst,” while other pro‑enforcement narratives emphasize raw enforcement volume and the presence of individuals with convictions. Both framings are supported by parts of the data but aim different political points—one to argue for restraint and oversight, the other to justify enforcement expansion. The source collection includes both government reporting and think‑tank analysis, signaling varied agendas [4] [3].
6. What is omitted or uncertain in the available reports
Available summaries do not uniformly report demographic details, case outcomes, or reasons for detention (e.g., immigration status, prior deportation orders, or public safety assessments). Absent standardized breakdowns—by charge type, conviction severity, or immigration status—readers cannot fully assess the security risk posed by the detainee population. Additionally, rapid changes in enforcement priorities create volatility in short‑term statistics, complicating comparisons to earlier administrations [5] [1] [3].
7. Bottom line for the original question: a qualified numeric answer
A straightforward, defensible statement is: At a late‑September 2025 point in time, roughly 27–33% of people in ICE detention had no criminal conviction, while broader arrest counts covering months show substantially higher shares (up to 65% in one analysis) depending on methods and windows used. Thus the percentage varies by dataset and definition, but multiple independent figures converge on the conclusion that a substantial minority—and under some measures a majority of arrests—are people without criminal convictions [1] [2] [3].