Call recording and retention for collections firms

Nothing in Regulation F requires a firm to record calls. If you do record, 12 CFR 1006.100(b) requires you to keep each recording for three years after the date of the call — a different clock from the general rule in 1006.100(a), which runs to three years after the last collection activity on the debt. Consent is a separate question and it is state law. Federal law permits recording where one party consents, and New York's eavesdropping offences turn on the absence of a party's consent; Florida makes interception lawful only "when all of the parties to the communication have given prior consent to such interception". Because recordings are evidence of compliance, a CFPB examiner is directed to listen to a sample of collection calls, and a bank's oversight review of a placement portfolio asks for them on the same footing.

The rule does not make you record. It makes you keep what you recorded.

Comment 1006.100(b)-1 says it in one line: "Nothing in § 1006.100 requires a debt collector to record telephone calls." What follows is the part that matters. "However, if a debt collector records telephone calls, the recordings are evidence of compliance or noncompliance with the FDCPA and this part, and, under § 1006.100(b), the debt collector must retain the recording of each such telephone call for three years after the date of the call."

The regulation itself is section 1006.100(b): "If a debt collector records telephone calls made in connection with the collection of a debt, the debt collector must retain the recording of each such telephone call for three years after the date of the call." There is no sampling allowance and no exception for calls that produced nothing. Each call.

The decision to record is therefore a decision to take on a retention obligation across the whole calling operation, and it should be made once, deliberately, rather than arrived at because a telephony platform shipped with recording enabled. Partial recording — some campaigns, some agents — is permissible, but it produces a record in which the absence of a recording proves nothing, which is the worst position to be in when a consumer describes a call you did not capture.

Two clocks, and they do not start together

The general rule in section 1006.100(a) is anchored to the account: "Except as provided in paragraph (b) of this section, a debt collector must retain records that are evidence of compliance or noncompliance with the FDCPA and this part starting on the date that the debt collector begins collection activity on a debt until three years after the debt collector's last collection activity on the debt." Recordings are carved out of that by the opening words and governed by 1006.100(b) instead — the section-level commentary states the two anchors side by side, requiring retention until three years after the last collection activity "or, in the case of telephone call recordings, until three years after the dates of the telephone calls."

That difference is not academic on a litigated file. An account worked for four years generates a recording in month two whose three-year minimum expires while the matter is still live and the account-level records still have to be kept. A purge job keyed to the account clock will keep it; a purge job keyed correctly to the call date will delete it. Both are defensible readings of the minimum, and only one of them leaves you with the evidence.

The rule sets a floor, and says so. Comment 1006.100-1 provides that "Nothing in § 1006.100 prohibits a debt collector from retaining records that are evidence of compliance or noncompliance with the FDCPA and this part for more than three years after the applicable date." Three years is also less than some of the exposure. An FDCPA action may be brought "within one year from the date on which the violation occurs" under 15 U.S.C. 1692k(d), which is shorter — but the FDCPA is not the only claim a recording answers, and a bank's own oversight cycle may reach back further than the regulation requires you to keep.

Consent is state law, and the states do not agree

Federal law sets a one-party floor. Under 18 U.S.C. 2511(2)(d), "It shall not be unlawful under this chapter for a person not acting under color of law to intercept a wire, oral, or electronic communication where such person is a party to the communication or where one of the parties to the communication has given prior consent to such interception" — subject to the exception for interception made for a criminal or tortious purpose. States may and do go further.

Florida goes further. Section 934.03(1)(a), Florida Statutes, makes it an offence to "Intentionally intercept, endeavor to intercept, or procure any other person to intercept or endeavor to intercept any wire, oral, or electronic communication", and the consent exception at section 934.03(2)(d) is written for all parties: it is lawful "for a person to intercept a wire, oral, or electronic communication when all of the parties to the communication have given prior consent to such interception." One party's consent is not enough, and the party whose consent is missing is the consumer.

New York builds the consent question into the definition of the offence. Section 250.05 of the Penal Law makes eavesdropping a class E felony, and section 250.00 defines its components by reference to consent: wiretapping is the intentional overhearing or recording of a telephonic communication by a person other than a sender or receiver "without the consent of either the sender or receiver", and mechanical overhearing is the intentional overhearing or recording of a conversation "without the consent of at least one party thereto, by a person not present thereat".

We name only the states whose own code we opened for this page. At the time of review, the official statute sites for Texas and Georgia did not serve statutory text to a plain HTTP request — both return a JavaScript shell — so no claim is made here about either. Editorial policy does not allow us to substitute a commercial mirror to fill that gap.

What a bank or an examiner can demand, and when

The CFPB's debt collection examination procedures list "Telephone recordings" among the documents an examiner requests at the outset, and are explicit about how they are used: "In assessing an entity's communications, examiners should consider documents and recordings, including hard copies or electronic copies of letters, voice recordings of telephone communications, and notes made during or after telephone calls or personal visits. Examiners should review a sample of records and listen to a sample of collection calls." The retention question is asked directly as well — the examiner is directed to "Determine whether the entity retains any recordings of telephone calls in connection with the collection of a debt for three years after the date of the call."

A bank supervising a placement portfolio asks a narrower version of the same question, usually against a sample of accounts and usually with a deadline measured in days. What determines whether that is a query or a project is not whether the recordings exist; it is whether they can be located by account and debt rather than by agent extension and timestamp.

Regulation F anticipates that the recordings may not be in your building. Comment 1006.100(a)-3 provides that "Records may be retained by any method that reproduces the records accurately (including computer programs) and that ensures that the debt collector can easily access the records (including a contractual right to access records possessed by another entity)." A contractual right to a vendor's recordings satisfies the rule — but only if the right is actually in the contract, survives termination, and covers the full retention period rather than the vendor's default storage tier.

What has to sit alongside the audio

A recording that cannot be tied to a debt is not evidence of anything. The metadata that makes it useful is the same metadata the frequency rule needs: which debt the call was placed against, the connection disposition, whether a conversation occurred, and the timestamp. If those live in the telephony platform and the account lives in the case management system, every request becomes a reconciliation.

Regulation F does not require you to build records you would not otherwise build. Comment 1006.100(a)-2 says a collector "need not create and maintain additional records, for the sole purpose of evidencing compliance, that the debt collector would not have created in the ordinary course of its business in the absence of the record retention requirement". But it also makes clear that what you do keep is in scope, and comment 1006.100(a)-1 sets the standard the whole exercise is measured against: the firm must retain records evidencing both that it did what the rule required and "that the debt collector refrained from conduct prohibited by the FDCPA and this part."

That second half is the reason retention design is a compliance decision rather than an infrastructure one. Proving that a call was never made during a suppression window requires the log covering that window to still exist. A purge that is correct to the minimum can still delete the only proof you had.

This is an informational reference, not legal advice, and using it creates no attorney-client relationship. Limitations periods turn on facts this page cannot know — which state's law governs, the contract type, when the claim accrued, and whether anything tolled or revived it. Confirm against the primary source and your own counsel before acting.