Percentage of ice detainees who have no criminal history
Executive summary
As of late 2025 and early 2026 independent analyses of ICE point-in-time detention counts show that a substantial majority of people in ICE custody had no criminal conviction on record—most sources put that share roughly between about 69 percent and 74 percent for detainees arrested by ICE (not including all border arrests) [1] [2] [3]. Broader tallies that combine ICE and CBP custody or that measure “arrests” rather than the in‑custody population produce lower figures—roughly half—because Border Patrol apprehensions (which tend to lack U.S. criminal histories) are treated differently in the datasets [4] [5].
1. What the numbers say: point-in-time detention counts show most detainees lack convictions
Nonpartisan trackers and policy analysts using ICE’s point-in-time detention snapshots reported that roughly 69–73.6 percent of people in ICE custody had no criminal conviction as of fall 2025 into early January 2026, with TRAC and Cato Group analyses citing about 69–72 percent and The World Data reporting 73.6 percent for early January 2026 [1] [3] [2]. These figures refer to people in ICE custody at those snapshot dates and are drawn from ICE’s published detention data, which classify detainees by criminal-history categories [5] [6].
2. Why different studies produce different percentages: definitions and populations matter
Estimates diverge because researchers measure different populations—“people detained after an ICE arrest,” the whole detained population including Border Patrol transfers, or all DHS arrests—and because they distinguish convictions from pending charges or immigration-related offenses; one analysis noted that nearly half (48 percent) of the total detained population (ICE plus CBP) lacked any criminal history on record, illustrating how inclusion of Border Patrol cases lowers the share of convicted detainees [4] [5]. DHS and ICE sometimes report arrest‑level statistics that emphasize arrests of people with charges or convictions, which applies to the subset of arrests rather than the custody population, creating an apples‑to‑oranges comparison highlighted by The World Data report [2].
3. The trend: a rapid shift toward detaining people without convictions in 2025
Multiple organizations documented a dramatic increase in detention of people without criminal convictions during 2025: analyses cited surges in non‑criminal detainees—sometimes expressed as multi‑hundred‑percent increases—and attributed much of the overall growth in ICE detention to people without criminal convictions, driving the detained population to record heights by year’s end [7] [8] [1]. Reports contend that a large fraction of the rise in daily bed counts since early 2025 comes from individuals with no criminal convictions, while those with convictions accounted for a much smaller share of the growth [4] [1].
4. Official framing and counterclaims: DHS/ICE emphasize criminal threats, but data nuance that claim
DHS and ICE officials have underscored arrests of serious criminals in publicity around enforcement actions, and DHS released lists of high‑profile criminal arrests to justify interior enforcement priorities [9]. Independent data reviewers note, however, that those narratives do not reconcile easily with point‑in‑custody counts showing most detainees lacked convictions; DHS has sometimes said a claimed percentage applies to arrests rather than the custody roster, a distinction that complicates direct comparisons [2] [3].
5. Limitations and remaining uncertainties in the public record
Available counts are useful but imperfect: ICE’s public datasets are point‑in‑time and periodically updated, categories mix immigration offenses and criminal convictions, and third‑party compilations vary in methodology; academic and advocacy groups warn that some DHS data releases and press statements conflate arrests, pending charges, and convictions, making precise, single‑number answers fraught without careful definition [5] [10]. Given those constraints, the most defensible, evidence‑backed answer is a range rather than an exact decimal.
6. Direct answer
Using ICE point‑in‑time detention data analyzed by TRAC, Cato, Migration Policy and independent data projects, the share of people in ICE custody with no criminal conviction was roughly 69–74 percent in late 2025 to early January 2026 for the ICE‑detained population analyzed by those groups; when broader DHS custody pools or mixed ICE+CBP totals are used, the fraction lacking criminal history falls toward roughly half (~48 percent), reflecting inclusion of Border Patrol apprehensions and different counting methods [3] [1] [2] [4].