How did ICE removals and arrest metrics change between 2013 and 2016, and what counting rules affected those totals?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
Between FY2013 and FY2016 reported ICE removals fell substantially in aggregate terms — from 363,144 removals in FY2013 (excluding a reporting “lag”) to a total of roughly 240,255 removals in FY2016 as counted by ICE (which includes both removals and returns) [1] [2]. That decline was shaped not only by operational emphasis on criminal priorities but also by several counting-rule changes and interagency processing shifts — notably an October 5 “lock” date, a June 2013 treatment of voluntary returns, and the growing role of Border Patrol-processed cases being credited to ICE — each of which altered the composition and year-to-year comparability of the totals [1] [3] [4] [2].
1. The headline numbers: fewer total removals, sharper interior decline
ICE’s internal reports show a drop from 363,144 removals in FY2013 (excluding the FY2012 “lag”) to roughly 240,255 total removals in FY2016 when removals and returns are combined [1] [2], and independent analysis highlights that only about 65,332 of those 2016 removals originated from the U.S. interior — a figure CIS reports as a 6 percent decline from 2015 and a much larger fall relative to earlier years [2].
2. Operational priorities reshaped who was removed
ICE’s Enforcement and Removal Operations (ERO) repeatedly emphasized prioritizing convicted criminals and threats to public safety, and by FY2016 ICE reported that 99.3 percent of its removals were CIEP-priority individuals with 83.7 percent classified as Priority 1 — a shift toward criminal-case removals that changed the mix of cases counted as removals [1].
3. Counting rules and administrative “locks” that distort year-to-year comparisons
ICE has used an October 5 “lock” for removals since FY2009, assigning removals confirmed after that date to the next fiscal year to ensure confirmed departures are counted in a single year; ICE reiterated that practice in its FY2015–FY2017 materials and used similar language about excluding “lag” years when reporting comparative totals [1] [5] [4]. That procedural cutoff means annual totals can shift because of when returns are confirmed, not only because of enforcement intensity [1].
4. Policy changes in 2013 and handoffs between CBP and ICE changed the ledger
Beginning June 1, 2013, ICE stated that voluntary returns recorded without an ICE intake case were not recorded as ICE removals, shifting many expedited or Border Patrol-processed departures outside ICE’s removal count [3] [4]. Simultaneously, Border Patrol processing and handover practices meant an increasing share of removals credited to ICE were in practice border-origin cases handled first by CBP, a dynamic analysts say makes historical comparisons to earlier administrations problematic [2].
5. Metrics, duplication, and data gaps that complicate interpretation
DHS monthly data and ICE releases count “immigration events” and can count the same person multiple times across encounters, arrests, and other metrics, and external reviewers have flagged undercounts or inconsistent treatments in ICE’s public reporting — for example, GAO found ICE understates detention totals by excluding some temporary-facility bookings, and OHSS/DHS notes the public tables count events not unique people [6] [7] [8]. Researchers and FOIA-based projects also warn that archived ICE files and dataset boundaries (notably a 2013 dataset break) affect trend lines and comparability across years [9] [10].
6. Competing narratives and what each side emphasizes
ICE’s narrative during this period emphasized a narrowed, criminal-focused enforcement effort and pointed to high proportions of priority removals as evidence of focused public-safety enforcement [1]. Advocacy and research groups highlight the steep fall in interior removals and question whether counting changes, CBP-to-ICE handoffs, and non‑ICE-logged voluntary returns obscure a real reduction in interior enforcement; CIS’s analysis stresses the interior-removal decline and warns about attribution of border cases to ICE [2] [11]. Both perspectives are supported by ICE’s public reports and by outside data reviews — but reconciliation requires careful attention to the counting rules and dataset breaks noted above [1] [3] [7].