How do counting methods for 'removals' and 'encounters' differ across DHS and ICE reports, and why does that matter?
Executive summary
DHS’s Office of Homeland Security Statistics (OHSS) treats “encounters” and post-encounter outcomes as linked immigration events and distinguishes removals, returns, repatriations, and Title 42 expulsions with explicit operational definitions that attempt to match unique encounters to outcomes [1] [2]. ICE’s public statistics, by contrast, present removals and related metrics from its operational databases with different inclusion rules, reporting lags, and exclusions for certain staging or temporary facilities—differences that produce measurable divergences and matter for legal status, policy debates, and public trust [3] [4] [5].
1. How DHS/OHSS defines and counts “encounters” and outcomes
OHSS constructs a Persist dataset that starts from agency operational reports and treats the unit of measurement as immigration events—encounters, administrative arrests, or credible fear claims—then links those unique encounters to enforcement outcomes using a reliable event identifier (CIV_ID) and explicit categories for removals, returns, repatriations, and Title 42 expulsions [6] [2] [1]. OHSS explicitly defines “removal” as compulsory movement based on a final order and “return” or “repatriation” as administrative or voluntary movements not based on a removal order, and reports that it matches unique border encounters to associated enforcement outcomes to avoid double-counting [2] [7].
2. How ICE’s reporting approach and operational data differ from OHSS aggregation
ICE publishes Arrests, Removal, Detention and Alternatives statistics drawn from its ENFORCE and EARM systems and confirms its public tables are subject to quarter‑end locking and later revisions; ICE also excludes some initial book‑ins tied to temporary staging facilities from public book‑in counts, which alters comparability with DHS aggregated tables [3] [4] [5]. ICE’s focus is operational—tracking arrests, intakes, detentions, and removals within ERO—with definitions oriented to case processing and monitoring [3] [8]. OHSS notes that records may not match agency reporting due to methodological and “as‑of” date differences, implicitly acknowledging that ICE and CBP internal reporting choices affect totals [6] [9].
3. Where the counting diverges in practice (and why those differences show up)
Divergence arises from at least three technical choices: unit of analysis (unique encounter vs. agency case record), inclusion/exclusion rules (e.g., voluntary returns, Title 42 expulsions, CBP removals), and record linkage/cleansing (OHSS uses CIV_ID matching; ICE may remove administrative records or exclude temporary facility book‑ins) [2] [5] [6]. These operational determinations—what to call a “removal” versus a “return,” whether to count a CBP expulsion or an ICE‑executed removal, and how to treat rapid Title 42 expulsions—create numeric gaps that are documented by OHSS and flagged by third parties as methodological variation [2] [1] [9].
4. Why the differences matter: law, policy, and public narrative
The label matters: a “removal” normally produces legal bars to reentry and corresponds to final orders, whereas “returns” or “repatriations” may be voluntary or administrative and carry different legal consequences—so counting affects individuals’ immigration records and future enforcement options [2] [8]. At the policy level, aggregated counts feed public claims about enforcement intensity and resource allocation; inconsistent or non‑synchronous reporting can be weaponized by advocates and officials to claim surges or declines in deportations [10] [11]. From a transparency standpoint, reporting lags, excluded facility counts, and agency choices about what to publish complicate independent monitoring and scholarly estimates, as noted by GAO and policy analysts [4] [11].
5. Competing narratives, hidden incentives, and limits in the record
Agencies have incentives to present data that align with operational priorities—CBP emphasizes reductions in “getaways” and removals/returns since policy changes, ICE highlights monitoring and integrity of its datasets while acknowledging revisions, and OHSS frames its KHSM as the most comprehensive cross‑agency aggregation but warns of methodological limits [10] [3] [1]. Advocacy organizations point to enforcement tactics and expanded surveillance as context for increases in interior arrests, which is a framing that relies on different evidence than agency tallies and speaks to political and moral stakes in counting choices [12]. Publicly available reporting does not resolve every inconsistency: OHSS, ICE, CBP, and outside analysts each document gaps or caveats in their methodologies, and some differences cannot be reconciled without access to line‑level records beyond what is published [6] [9].
6. Bottom line — how to interpret headline totals responsibly
Treat headline counts as method‑dependent: DHS/OHSS totals reflect linked encounters across agencies with explicit definitions for removals versus returns, while ICE’s operational statistics reflect its internal case processing choices, exclusions, and reporting cadence—misreading the two as identical will produce incorrect inferences about legal outcomes and enforcement intensity [2] [3] [6]. Analysts should check which universe (CBP encounters, ICE book‑ins, OHSS Persist events) underlies any cited number and look for caveats about exclusions, Title 42 expulsions, and temporary‑facility book‑ins before drawing policy conclusions [2] [5] [1].