How does ICE classify 'criminal' versus 'noncriminal' detainees in its public statistics?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
ICE’s public statistics separate detainees into buckets that are not a simple binary of “criminal” versus “noncriminal”: the agency reports people with criminal convictions, people with pending criminal charges, and those without convictions or charges (often labeled “immigration violators” or “other immigration violators”), and it also assigns internal “threat levels” that can leave many detainees uncategorized (no ICE threat level) [1] [2]. Independent analyses and news reporting show those labels matter politically and practically, because they mix immigration-only offenses (civil violations like unlawful presence or reentry) with a range of criminal histories and pending allegations in ways that can overstate or understate public‑safety risk depending on how the numbers are presented [3] [4] [5].
1. How ICE’s public buckets are defined — not a simple criminal/noncriminal split
ICE’s online statistics break the detained population into groups that include people with criminal convictions, people with pending criminal charges, and a third group described as individuals “without any known criminal convictions or pending charges” who may nevertheless be held for immigration-specific reasons such as unlawful reentry after deportation, being an immigration fugitive with a final removal order, or foreign fugitives wanted abroad; ICE explicitly treats those immigration violations as civil matters even when detention results [1]. Public summaries therefore present three operational categories rather than a pure dichotomy, and ICE says custody decisions are individualized, considering criminal history, immigration history, family ties and humanitarian factors [1].
2. How third parties and press translate ICE categories into “criminal” vs “noncriminal”
Advocates, think tanks and news outlets typically collapse ICE’s tripartite taxonomy into a two‑part narrative by labeling the third group “noncriminal” or “immigration violators,” which produces headlines like “immigrants with no criminal record now largest group in ICE detention” (The Guardian) and analyses showing a majority of detainees lack a criminal conviction [4] [6]. Researchers such as Cato and news organizations like CBS and Fortune have highlighted that a large share of detainees have no recorded criminal conviction; Cato’s analysis and aggregated trackers reported figures in which roughly 65–73% of people in ICE custody lacked criminal convictions at certain snapshots [7] [6] [8] [2].
3. The complication of pending charges, reentries and immigration-only offenses
ICE and outside analysts diverge when counting people with pending charges: ICE sometimes designates arrests with pending criminal allegations under a criminal umbrella in public statements, while critics note many of those charges are unresolved, dismissed, or relate to minor offenses [3]. Likewise, ICE’s “no convictions or pending charges” category can include serious immigration-related conduct — repeated illegal reentry, fugitives with final removal orders, or foreign‑based criminal warrants — meaning “no conviction” does not automatically equal “low risk” or innocent of all legal violations [1].
4. Threat-level coding and missing detail on severity of offenses
Beyond convictions and charges, ICE assigns internal threat levels (1–3) to detainees, but many detainees are listed as having “no ICE threat level,” and publicly posted headcounts typically do not break down the seriousness of convictions or the nature of pending charges [2] [8]. Major news reporting notes that publicly released tallies rarely specify whether convictions are violent felonies, misdemeanors, or immigration-related crimes, which leaves significant ambiguity about the public‑safety implications of the detainee population [8] [5].
5. Politics, framing and the incentives in the data
How ICE’s categories are described is politically freighted: administrations and DHS officials emphasize “criminal illegal aliens” to justify expanded detention capacity, while critics and watchdogs stress the growing share of people held for civil immigration violations and the costs of detaining people without criminal convictions [5] [7]. Independent trackers and think tanks using ICE’s public data can produce divergent emphases—either that most detainees lack criminal convictions (Cato, TRAC, The Guardian) or that sizable minorities have charges or convictions—because the underlying ICE categories allow multiple, defensible framings [6] [7] [4].
6. Bottom line and limits of public reporting
ICE classifies detainees using multiple labels—convicted, charged, and those without convictions or charges (including certain immigration fugitives and re‑entries)—and assigns threat levels that are imperfectly disclosed; the agency’s data therefore do not map cleanly to a legal or moral binary of “criminal” versus “noncriminal,” and independent analyses relying on ICE’s public feeds must make interpretive choices about pending charges, immigration offenses, and threat coding that affect reported proportions [1] [3] [2]. Where sources disagree, it is often over categorization choices (pending charges vs. convictions) and the omission of offense severity in public tallies [8] [5].