How does ICE define and categorize 'criminality' in its public detention and removal tables?
Executive summary
ICE’s public detention and removal tables classify “criminality” along a few discrete, administratively defined lines: persons with criminal convictions, those with pending criminal charges, and those with no convictions or pending charges but who are considered immigration violators (including repeat re‑entrants, immigration fugitives, and foreign fugitives) [1]. Those categories are further parsed in federal statistical products that report convictions by severity—misdemeanor, felony, and aggravated felony—and by arresting agency or custodial pathway, but ICE and DHS sources warn the underlying records and definitions change over time and are subject to data limitations [2] [3].
1. What ICE officially lists as “criminality” in its public tables
ICE’s Enforcement and Removal Operations (ERO) statistics present criminality primarily by three administrative buckets: convicted offenses, pending criminal charges, and “no convictions or pending charges” which nevertheless can include repeat re‑entry after deportation, immigration fugitives with final removal orders, and foreign fugitives wanted abroad [1]. The agency’s public tables also record whether a detainee had a prior conviction at booking and link back to arresting agencies—ICE, Customs and Border Protection, or custodial arrests originating in local, state or federal lockups—so “criminality” in the dataset is a mix of legal status (convicted/pending) and operational source (custodial pathway) [1] [4].
2. How conviction severity and offense types appear in the statistics
DHS statistical products and third‑party summaries break convicted categories down by severity: for example, recent reports show detainees with criminal records have been categorized as misdemeanors, felonies, and aggravated felonies, with published shares for each category [2]. Independent analyses citing ICE and DHS data have quantified the share of detainees with convictions versus those without, noting that a sizeable fraction of people in ICE custody have no criminal conviction while convicted populations include a mix of misdemeanors, felonies and a smaller share of violent or aggravated felony convictions [2] [5].
3. Administrative classification choices and data practice caveats
ICE and DHS emphasize the data come from live case management systems and are subject to changes, directed administrative record removals, and reclassification; the Office of Homeland Security Statistics notes records are not de‑duplicated and can be corrected, so snapshots reflect evolving administrative choices [3]. Analysts have highlighted additional operational classification changes—such as ICE’s consolidation of local, state and federal “Criminal Alien Program” arrest pathways into a generic “Custodial Arrests” label—which can shift how criminality appears in the public tables without an underlying change in who is being detained [6]. Several watchdog and policy groups have also pointed out that ICE’s public labels (convicted/pending/none) do not disclose offense details consistently across datasets, and ICE’s arrest tables sometimes exclude Border Patrol arrests or federal immigration‑related criminal charges, complicating direct comparisons [7] [4].
4. What the categories do—and don’t—say about targeting and public safety
The numerical categories in ICE’s tables are administrative, not normative: a “no conviction” label in ICE data does not mean the person posed no security risk according to ICE’s internal case reviews, and conversely a conviction label encompasses a wide range of offense severity from minor misdemeanors to aggravated felonies [1] [2]. Independent researchers and advocacy groups using the same public tables have drawn contrasting conclusions—some argue the tables show the agency is detaining many non‑criminal immigration violators, while others point to custodial arrest pathways and prior convictions as evidence of reliance on criminal‑justice referrals—underscoring that interpretation depends on how one weights conviction severity, arrest source, and the legal definition of “removable” [5] [6] [4].
5. Transparency gaps and what remains uncertain in the public record
Public ICE tables document categories and counts, but they stop short of providing a uniform, granular mapping of offense codes, duration since conviction, or the agency’s internal rationale for classifying an individual as a criminal priority; DHS warns the data are subject to change and administrative filtering, and independent analysts note missing disclosures about how many custodial arrests originate from local jails versus at‑large interior operations [3] [6]. Where the public data lack detail, outside researchers have used DHS/OHSS extracts and FOIA‑obtained records to approximate patterns, but those efforts reveal that conclusions about “who” ICE detains hinge on definitional and reporting choices embedded in ICE’s public tables [3] [5].