How have other states reconciled DHS detainer counts with state and local jail records since 2020?
Executive summary
States and localities have used a mix of data-matching, policy changes, and legal frameworks to reconcile DHS/ICE detainer counts with jail records since 2020, but practices and transparency vary widely and no single national protocol has emerged [1] [2]. Data mismatches are driven by differing reporting windows, definitions (detainer request vs. ICE custody), and local sanctuary policies that limit honoring detainers, and advocacy groups and academic projects have stepped in to bridge the gaps [2] [3] [1].
1. How the raw numbers diverge: reporting systems and timing
The first and most consistent reason counts don’t line up is methodological: DHS and ICE publish detainer and detention figures from federal operational systems that use validated but agency-specific identifiers and "as-of" dates, while state and local jails use their own booking and release logs, leading to predictable mismatches unless reconciled through record-level linking [2] [1]. ICE’s statistical systems de-duplicate and standardize component reports and acknowledge that records “may not match agency reporting due to differences in reporting methodologies or as-of dates,” meaning a jail’s short-term hold may never appear as an ICE detention even if an ICE detainer was issued [2]. The Deportation Data Project has shown that the only reliable way to reconcile is by joining detainers to arrests and detention tables at the individual-record level using unique identifiers—work that outside researchers have undertaken because public summaries obscure those linkages [1].
2. Local policy decisions: sanctuary rules and declined detainers
A second mechanism is policy: many sanctuary jurisdictions have formal prohibitions on honoring ICE detainers or require judicial warrants, which means state or county jails will decline to hold someone for ICE and therefore the federal detainer count may overstate the number of people actually transferred to ICE custody [3]. ICE itself documents that it records detainers and updates records when a detainer is declined, creating an official federal trail of requests that does not equate to transfers to federal custody [4]. Reconciling the two data sources therefore often requires states to publish not only bookings but whether a detainer was honored, declined, or resulted in transfer—data that some localities release and others do not [1] [4].
3. Technical reconciliation: matching fingerprints, identifiers, and timestamps
Where reconciliation has been done rigorously, researchers and some jurisdictions have used biometric and booking identifiers—fingerprints, booking numbers, and timestamps—to align ICE detainer requests with local jail records and subsequent ICE book-ins, allowing clear pathways from detainer to detention or non-honoring [1]. The Deportation Data Project’s individual-level datasets demonstrate this approach, and DHS’s own statistical systems rely on unique record identifiers to deduplicate and standardize reporting, but that alignment requires data-sharing agreements or public data dumps that are not universally available [1] [2].
4. Accountability via independent projects and advocacy reporting
Because official transparency is limited, independent projects and advocacy groups have become de facto auditors: organizations like the Deportation Data Project and policy researchers have combined ICE data releases with local booking records to estimate how many detainer requests led to transfers, and they have flagged duplicate records and reporting inconsistencies that government summaries miss [1] [5]. These outside analyses often carry an implicit agenda—advocates seek to show the effects of sanctuary or federal enforcement—so their work must be read alongside agency documentation that explains methodological differences [5] [2].
5. Legal and political constraints shaping reconciliation
Courts and state law also shape how reconciliation plays out: litigation restricting DHS’s ability to take custody upon issuance of detainers in some jurisdictions, and state laws banning or limiting honors of detainers, change the ground truth that data analysts must reconcile—what looks like a federal detainer on paper may legally be unenforceable locally, a reality emphasized in legal overviews of detention authority [6] [3]. That legal fragmentation means a national reconciliation requires not just technical matching but mapping which jurisdictions will or will not hand people over to federal custody [6].
6. What reporting gaps remain and where reconciliation is incomplete
Public sources make clear how to reconcile in theory—match unique IDs, include disposition codes, disclose honor/decline status—but they also make clear that many jurisdictions do not publish the necessary granular fields and DHS summaries use agency-specific formats, so comprehensive, nationwide reconciliation since 2020 remains incomplete and largely the work of outside data projects and targeted state audits [2] [1]. Where claims are made about large discrepancies, those claims must be checked against record-level joins or jurisdictional disclosures because summary statistics alone cannot determine whether a detainer was honored, declined, or led to federal custody [1] [4].