Information a State Shares With Federal Databases

The belief that state government is continuously feeding information to federal immigration authorities is almost universal and almost entirely wrong. What exists instead is a handful of specific channels, each created by a particular statute for a particular purpose, with defined participants and defined content. Everything else is either a query that somebody asked, or nothing at all.

The criminal history channel

This is the largest and most automatic of the routes, and it is worth understanding in detail because it accounts for most of what actually travels.

The mechanism is fingerprints. When a person is booked into custody, fingerprints are taken and submitted to the state's central criminal history repository. The repository forwards them to the national identification system maintained federally. The record is created at that moment, and it is created by the act of booking rather than by any decision about the case.

The record is available to authorized users. Federal agencies with statutory access can query the national system, and immigration authorities are among them. This is a query against an existing record rather than a notification pushed to them, but the practical difference is small where a query is made routinely.

Dispositions are reported separately and imperfectly. Once a case concludes, the court is supposed to report the outcome back into the state repository, which updates the national record. Compliance is uneven, and a substantial share of records nationally show an arrest with no recorded outcome. That incompleteness works against the individual, because an adjudicator reading the record sees a charge and no acquittal, dismissal or reduction.

The remedy is documentary. Where a disposition has not propagated, the fix is a certified copy of the court record rather than a request to update the database. That is faster, it is authoritative, and it is the form the adjudicator wants. How such a record is read once it arrives is the subject of how a state disposition reads to a federal adjudicator.

Driver and vehicle systems

The second channel is narrower than reputation suggests and is governed by its own restrictions.

State-to-state sharing exists for licensing integrity. States exchange information to prevent a person holding licenses in several states at once and to ensure that a suspension in one state is visible in another. The systems that do this are operated among the states themselves for that purpose.

Federal privacy law restricts disclosure of driver data. Personal information from motor vehicle records is protected by statute, with an enumerated list of permissible uses. The list includes use by government agencies carrying out their functions, which is a real exception, but it is an exception within a restrictive framework rather than an open channel.

States that issue cards without a status requirement legislated further. Many of them enacted confidentiality provisions covering the application information specifically, exempting it from public records requests and restricting its use. Those provisions bind state agencies and employees; their reach is discussed in what a state non-cooperation statute actually limits.

The federal identity card standard has its own retention rule. Where a state issues cards meeting the federal minimum standard, it retains images of source documents for a defined period. That is a state-held record subject to state law, not a transmission, and the mechanics are described in what a compliant card requires.

Queries rather than feeds

The third category is the one most often mistaken for a feed, and the distinction matters.

Verification services answer questions. A state agency administering a program with a status-based eligibility rule submits identifying details about an applicant and receives a response. Nothing is transferred beyond what the query needs, and the query is generated because a program rule required it.

Employment verification works the same way. An employer checks a specific person's authorization against federal records. The employer initiates it; no state agency is involved; and the result concerns employment eligibility.

Queries leave a trace at the asking end. The agency that asked records the answer in the applicant's file, subject to that program's safeguarding rules. What those rules permit is set out in what a benefits agency reports and to whom.

No query means no flow. This is the practical point. Where a program has no status-based eligibility rule, there is no reason for a query and none is made. A person using a service that does not condition eligibility on status has not generated a lookup by using it.

What sits outside entirely

Several categories of record that people worry about are not connected to any of the above.

Education records. Protected federally, disclosable without consent only in defined circumstances, and not part of any immigration data flow. The tuition affidavit discussed in the affidavit a student signs lives inside this regime.

Health information. Governed by its own confidentiality framework, with its own permitted disclosures. Treatment records are not queried in the ordinary course of an adjudication.

Benefit program records. Subject to safeguarding provisions restricting use to program administration, audits, other benefit programs and lawful process.

Ordinary civil court files. Public in most states and therefore available to anyone who looks, which is a different proposition from being fed anywhere. A public record is discoverable; it is not transmitted.

The honest summary is that the flows are narrower and more specific than the general anxiety assumes, and that the largest one is triggered by an arrest rather than by contact with civil government. Where a person's own history includes something that could matter, the useful step is to obtain the record and read it rather than to guess at what a system might hold — and where the record is ambiguous or incomplete, having it reviewed by an immigration lawyer who reviews records disclosure before any filing is made is considerably cheaper than discovering the problem in an adjudication.

One structural point ties the whole picture together and is worth stating explicitly, because it is the difference between a useful mental model and a vague dread. The channels described above were each built for a purpose that has nothing to do with immigration: identifying repeat offenders, preventing a suspended driver from licensing in a second state, stopping duplicate benefit enrollment, confirming that an applicant meets a program's own eligibility rule. Immigration authorities are users of some of these systems, in the same way other federal and state agencies are users, and the systems behave the same way whoever queries them. Nothing was designed as an immigration surveillance apparatus, which is why the coverage is patchy, the data is frequently stale, and whole categories of ordinary life — school, healthcare, employment history, civil litigation — sit entirely outside it. The practical consequence is that the record which matters in almost every case is the criminal history record, and that record is obtainable by the person it concerns. Reading one's own file is the single most useful thing available, and it costs a form and a fee. What that record then means in a federal analysis is a separate exercise set out in how a state disposition reads to a federal adjudicator.