Reassigned Numbers Database

A number a subscriber gives up does not stay out of service. The carrier holds it, then assigns it to someone new. Consent captured on a lead belongs to the person who held that number at the time, not to whoever holds it now, and nothing audible in the call reveals the swap. The FCC's Reassigned Numbers Database exists to make that swap visible before the phone rings.

The one question the database answers

The query is not who owns this number. Under 47 CFR 64.1200(m)(1) a caller submits the number together with the date prior express consent was obtained, and the database reports whether the number has been permanently disconnected since that date. A response of no means no permanent disconnection was recorded in that window. The data behind it comes from carriers: paragraph (l) requires reporting carriers to keep the most recent permanent disconnection date for each North American Numbering Plan number allocated or ported to them, and to file it monthly.

Paragraph (l)(3) draws the boundary. A number counts as permanently disconnected when the subscriber has relinquished it, or when the carrier has reversed the assignment so the number is no longer associated with that subscriber. A number ported to another carrier is expressly not permanently disconnected, so a port is not the event the database records.

The safe harbor turns on the database being wrong

Paragraph (m) says a person will not be liable for violating the prohibitions in paragraph (a)(1), (2) or (3) when the number had stopped belonging to the subscriber who gave consent. Two things have to be demonstrated, and the rule places the burden of proof and persuasion on the caller. The first is that the caller queried the database and received a response of no. The second is that the call happened because that no was erroneous.

The second condition is the whole shape of the protection. This is not a reward for diligence in general. It is cover for the specific case where the database was consulted and returned an answer that was not true.

Source: 47 CFR 64.1200, current text, read on 20 September 2026 through the eCFR renderer API at ecfr.gov. Paragraph (m) supports the conditional relief from the prohibitions in paragraph (a)(1), (2) and (3), the requirement that prior express consent was previously obtained, the caller's burden of proof and persuasion, the query consisting of the number and the consent date, the required response of no, and the requirement that the call resulted from the database erroneously returning that no. Paragraph (l) supports the carrier record keeping and the monthly report of permanent disconnection dates, and paragraph (l)(3) supports the definition of permanent disconnection and the statement that porting is not one.

Three gaps the safe harbor does not close

A caller who never had prior express consent takes nothing from paragraph (m). The text begins from a number for which consent was previously obtained, so the provision has nothing to operate on. A caller who skipped the query fails the first condition. A caller who received a response other than no and dialed anyway fails it too.

The third gap is one of scope. Paragraph (m) names only the prohibitions in paragraph (a)(1), (2) and (3), which restrict calls using an automatic telephone dialing system or an artificial or prerecorded voice. The do-not-call obligations of the same section sit in different paragraphs, and paragraph (m) does not mention them.

Where the rules are actually codified

Both halves live inside one section, 47 CFR 64.1200: paragraph (l) for the carrier reporting that fills the database, paragraph (m) for the safe harbor that uses it. There is no separate subpart of Part 64 carrying its own numbering for the database. Section 64.6400 and the sections after it, which look like a plausible home, belong to Subpart II on service protections for victims of domestic violence and human trafficking, and govern line separation from shared mobile contracts.

No figure appears on this page for how many numbers change hands in a year, because neither paragraph supplies one and no primary source read here does either. Counts in circulation come from companies that sell access to number data.

Not the same problem as a lead denying the form

Two complaints sound identical at hello. One is the consumer who never submitted anything and says so, which has its own set of causes and is covered in why leads say they never filled out a form. The other is the consumer who genuinely does not recognise the inquiry because it belongs to whoever held the number before them. In the second case the record is authentic and the person answering is simply not the person who signed it.

What that consent has to look like, and how it ends, are separate questions handled in prior express written consent and revocation of consent. Paragraph (m) takes valid consent as its starting point and speaks only to what happens to the number afterwards.

Nothing in paragraph (m) orders anyone to query. It creates a defence rather than a duty, and the defence is narrow: one erroneous no, proven by the caller who relied on it.

Source on the numbering: the section structure of Title 47 retrieved from the eCFR versioner API on 20 September 2026 places 64.6400 through 64.6409 in Part 64, Subpart II, Communications Service Protections for Victims of Domestic Violence, Human Trafficking, and Related Crimes, covering line separation requests rather than reassignment. An eCFR full-text search for the phrase Reassigned Numbers Database returns 47 CFR 64.1200; a search for the word reassigned also returns 47 CFR 52.103, which was opened and found to concern toll free number lag times, and is therefore not cited here. No other section of Title 47 was found to codify the database, so this page cites 64.1200(l) and (m) and nothing else.