
A $312 charge at a gas station outside Bakersfield sits four screens down in the transaction list. The account holder has never driven through Bakersfield. The charge posted on July 19, which put it on the statement the bank mailed on July 31, and today is October 12.
The card never left the wallet. Nothing was lost, nothing was reported stolen, and the money is gone anyway. What decides whether the bank puts that $312 back has almost nothing to do with the fraud itself. It turns on the distance between July 31 and the phone call.
Two numbers, and the date that picks between them
Regulation E governs electronic money leaving a consumer deposit account, and it does something unusual for a consumer protection rule: it caps the customer’s loss, then makes the cap conditional on speed. Section 1005.6 sets out the tiers.
The first one is the familiar figure. Notify the institution within two business days after learning of the loss or theft of an access device, and liability “shall not exceed the lesser of $50 or the amount of unauthorized” transfers that occurred before notice. That is the $50 everyone half-remembers from a card agreement.
Miss the two days and the ceiling moves. Under 1005.6(b)(2), liability runs up to $500, built from two pieces: “$50 or the amount of unauthorized transfers that occur within the two business days, whichever is less” plus “the amount of unauthorized transfers that occur after the close of two business days and before notice to the institution, provided the institution establishes that these transfers would not have occurred had the consumer notified the institution within that two-day period.”
That proviso is doing real work. The bank has to show the delay caused the later losses. A card that was cloned once and used once, with no further activity, does not produce a $500 exposure simply because the customer took a week to call.
Then comes the tier that has no ceiling at all. Section 1005.6(b)(3) reads: “A consumer must report an unauthorized electronic fund transfer that appears on a periodic statement within 60 days of the financial institution’s transmittal of the statement to avoid liability for subsequent transfers. If the consumer fails to do so, the consumer’s liability shall not exceed the amount of the unauthorized transfers that occur after the close of the 60 days and before notice to the institution.”
Read that last clause slowly, because it is a cap in grammar only. Everything after day 60 is on the account holder. There is no dollar figure attached to it.
| When notice reaches the institution | Ceiling on the account holder’s loss |
|---|---|
| Within two business days of learning a device was lost or stolen | Lesser of $50 or the unauthorized transfers before notice |
| After two business days, device lost or stolen | Up to $500, and only for transfers the institution shows the delay allowed |
| Within 60 days of the statement that first shows the transfer | The tiers above still govern |
| After 60 days from that statement | No cap on transfers occurring after day 60 and before notice |
The 60-day clock runs from transmittal of the statement, not from the day the charge posted and not from the day anybody noticed it. A paper statement that sat in a pile for three weeks burned three weeks of the window. So did an emailed statement notification filtered into a promotions folder.
There is one release valve. Under 1005.6(b)(4), “if the consumer’s delay in notifying the financial institution was due to extenuating circumstances, the institution shall extend the times specified above to a reasonable period.” Hospitalization and extended travel are the standard examples. The provision exists, it is written as a duty rather than a courtesy, and it is worth naming explicitly on the call rather than hoping the representative thinks of it.
One more piece of 1005.6 gets overlooked because it sits in paragraph (a). A consumer can be held liable at all “only if the financial institution has provided the disclosures required by § 1005.7(b)(1), (2), and (3).” Those are the initial disclosures covering liability, the phone number and address for reporting, and the institution’s business days. If they never arrived, the tiers above never attach.
What counts as unauthorized, and what quietly does not
Section 1005.2(m) defines the term as a transfer “from a consumer’s account initiated by a person other than the consumer without actual authority to initiate the transfer and from which the consumer receives no benefit.” Three carve-outs follow, and two of them catch ordinary households.
The first excludes a transfer “initiated by a person who was furnished the access device to the consumer’s account by the consumer, unless the consumer has notified the financial institution that transfers by that person are no longer authorized.” Hand a debit card to an adult child for a grocery run and the card is furnished. What that person spends afterward is outside the definition until the institution has been told the authority is revoked. The revocation is a call to the bank, not a conversation at the kitchen table.
The second excludes transfers made “with fraudulent intent by the consumer or any person acting in concert with the consumer.” The third excludes transfers by the institution or its own employee, which are handled elsewhere.
The load-bearing words in the main sentence are “initiated by a person other than the consumer.” A payment the account holder pushed out of the account personally was initiated by the account holder, whatever story preceded it. The regulation is written around transfers taken from an account, not around payments talked out of one. Anyone who sends money after a phone call from someone claiming to be the bank’s fraud department is standing outside that sentence, which is precisely why that script keeps working.
Section 1005.2(d) settles a smaller question that changes the arithmetic: a business day is “any day on which the offices of the consumer’s financial institution are open to the public for carrying on substantially all business functions.” A branch open Saturday morning for teller work only is not automatically a business day at that institution, and two business days over a holiday weekend can stretch to Wednesday.
What the bank owes once the notice lands
Section 1005.11 takes over at the moment notice arrives, and it is written as a schedule.
A notice of error is not a complaint form. Paragraph (b)(1) asks for three things: enough to identify the consumer’s name and account number, an indication of why the consumer believes an error exists, and the type, date, and amount of the error. The error itself can be more than fraud. The section covers an unauthorized transfer, an incorrect transfer in or out, an omission from a periodic statement, a computational or bookkeeping error by the institution, receipt of an incorrect amount of money from a terminal, a transfer not identified properly, and a request for documentation or clarification.
The notice must reach the institution “no later than 60 days after the institution sends the periodic statement or provides the passbook documentation, required by § 1005.9, on which the alleged error is first reflected.” Same 60 days, measured the same way.
From there the institution “shall investigate promptly and … shall determine whether an error occurred within 10 business days of receiving a notice of error.” If it finds one, it must correct it within one business day of that determination.
Ten business days is often not enough, and the rule anticipates that. The institution may take up to 45 days instead, on one condition: it provisionally credits the account “in the amount of the alleged error (including interest where applicable) within 10 business days.” It then has two business days to tell the consumer the amount and date of that credit, and it must give full use of the funds while the investigation continues. Where it has a reasonable basis for believing an unauthorized transfer occurred and has met the 1005.6(a) disclosure requirement, it may hold back a maximum of $50 from the provisional credit. The CFPB describes the same holdback in its plain-language summary: a temporary credit “for the amount of the disputed transaction, minus a maximum of $50, while it continues to investigate.”
Three situations stretch the schedule. The deadlines become 20 business days in place of 10, and 90 days in place of 45, for transfers involving an account in its first 30 days after the first deposit, for point-of-sale debit card transactions, and for transfers not initiated within a state.
| Circumstance | Determination due | Outer limit with provisional credit |
|---|---|---|
| Standard error notice | 10 business days | 45 days |
| Point-of-sale debit card transaction | 10 business days | 90 days |
| Transfer not initiated within a state | 10 business days | 90 days |
| Account less than 30 days past first deposit | 20 business days | 90 days |
The point-of-sale row is the one that surprises people, because it covers the most common kind of card fraud there is. A disputed swipe can legitimately sit under investigation for three months, with the disputed amount sitting in the account the whole time as provisional credit.
The written confirmation that decides whether credit appears
Oral notice starts the clock. Paragraph (b)(2) allows the institution to “require the consumer to give written confirmation of an error within 10 business days of an oral notice,” and paragraph (c)(2)(ii) ties the provisional credit to that confirmation.
The practical shape of it: a call on Monday opens the dispute, the representative mentions a form, and if the form does not come back inside ten business days, the institution is not obliged to put the money in the account while it investigates. The dispute survives. The temporary money does not. On a $312 charge that is an annoyance. On a drained payroll deposit it is the difference between paying rent that month and not.
Section 1005.6(b)(5)(ii) says the consumer “may notify the institution in person, by telephone, or in writing.” All three start the liability clock. Only one of them leaves a record with a date on it, which is the reason to follow a call with a secure message the same day even when nobody asks for it.
When the answer comes back no
Paragraph (d) covers the institution concluding that no error occurred, or that a different error occurred. It owes a written explanation of its findings and has to tell the consumer about the right to request the documents it relied on. That request is worth making. The documents are usually a terminal log and a transaction record, and they are the only way to see what the institution actually examined.
If provisional credit was given and is now being taken back, the institution has to notify the consumer of the date and amount of the debiting, and it must “honor checks, drafts, or similar instruments payable to third parties and preauthorized transfers from the consumer’s account (without charge to the consumer as a result of an overdraft) for five business days after the notification.” Five business days of cushion, then the account is on its own. Anyone who has been spending against a provisional credit for six weeks should treat that notice as a deadline rather than a letter.
Then there is paragraph (e), which is short and final. “A financial institution that has fully complied with the error resolution requirements has no further responsibilities under this section should the consumer later reassert the same error.” Calling back next month with the same claim and no new information does not restart anything. New evidence, a state regulator, or a court is what comes after a denial, not a second phone call.
Debit and credit are not the same instrument
The contrast is sharper than most card agreements let on. Under Regulation Z, liability for unauthorized use of a credit card “shall not exceed the lesser of $50 or the amount of money, property, labor, or services obtained by the unauthorized use before notification to the card issuer.”
There is no two-business-day tier in that sentence and no 60-day cliff attached to it. The credit card cap is a flat $50 that does not decay with delay. And when a credit card is compromised, the money in dispute belongs to the issuer until the dispute resolves. When a debit card is compromised, the money is already out of the account, and the account holder waits on a provisional credit to get it back.
For a household that keeps its cash in the account the card draws on, that difference lands during the exact weeks the balance is needed most. It is the strongest practical argument for not using a debit card on unfamiliar sites and at unattended terminals, and it has nothing to do with rewards.
What this changes on a normal Tuesday
- Find the transmittal date, not the posting date. The 60 days runs from when the statement went out. Pull the statement and write that date at the top of any note about the dispute, because every deadline in the file is measured from it.
- Open statements on a schedule. A monthly scan of the transaction list is the only mechanism that keeps a fraudulent charge inside the capped tiers. Nothing else in Regulation E does that work for you.
- Call first, then put it in writing the same day. Both start the clock. Only the written version proves when the clock started, and a written confirmation may be required within ten business days before provisional credit appears.
- Say “extenuating circumstances” out loud if they apply. Hospitalization or extended travel triggers a duty to extend the deadlines, not a discretionary favor.
- Revoke a furnished card at the bank. Taking the card back from a relative does not move those transactions inside the definition of unauthorized. Telling the institution the person is no longer authorized does.
- Ask for the documents after a denial. The written explanation comes with a right to the records the institution relied on, and paragraph (e) means a second unarmed phone call accomplishes nothing.
- Watch the five business days after a reversal. If provisional credit is debited back, third-party checks and preauthorized transfers are protected from overdraft charges for five business days and not a day longer.
The rule gives an account holder two chances to stay inside a cap: two business days from learning a card is gone, and 60 days from a statement. Both are counted by someone else, from a date that gets set whether or not the envelope is ever opened.
Further Reading
- What Happens to a Savings Account or CD When the Owner Dies
- The Savings Withdrawal Limit Was Repealed. Your Bank Kept It Anyway
- How to Verify a Bank Is Actually FDIC-Insured Before You Transfer
- When ‘FDIC Insured’ Means a Partner Bank, Not Your App
- Is Your Money Market Fund Safe? What Changed After 2023