What Household Debt Reports Miss About Buy Now, Pay Later

The Measurement Gap

Many pay-in-four Buy Now, Pay Later loans do not appear in credit reports, so credit-report-based household debt totals can miss some of them. The gap is real. Its size is not directly observed in the sources below, and annual BNPL lending cannot be added to a debt balance measured at one point in time.

The Federal Reserve Bank of New York’s Q3 2024 Household Debt and Credit release put household debt at $17.9 trillion at the end of September 2024. That is a dated historical anchor, not a live total. The report is based on the New York Fed Consumer Credit Panel, which is drawn from Equifax credit records.

The New York Fed’s Consumer Credit Panel FAQ says reported BNPL loans are likely included in the report’s “other” category, although many BNPL loans are not reported to credit bureaus. That supports a coverage limit. It does not support treating every BNPL loan as excluded.

Annual Lending Is Not Debt Outstanding

The Consumer Financial Protection Bureau’s December 2025 market report covers six large lenders. Those firms originated 335.8 million BNPL loans totaling $45.2 billion during 2023, after the CFPB adjusted the dollar figure for inflation. The agency says its sample is not necessarily representative of the whole market.

Those figures measure loans made during a year. They do not measure how much principal remained unpaid on one date. The CFPB also collected firm-level summaries rather than loan- or account-level records, so the same person can appear in more than one provider’s user total.

American Default’s Phantom Debt indicator follows the CFPB’s own nominal record for those same six lenders: $43.9 billion originated during 2023. That is the same collection as the $45.2 billion above, counted in the dollars of the year it was lent rather than adjusted to 2024 dollars. Either way it is a yearly flow, not an observed national stock of outstanding BNPL debt. The Richmond Fed separately models an average outstanding balance from assumptions about down payments, repayment cycles, and amortization. That estimate is a model, not a comprehensive balance reported by every provider.

Borrower Evidence Does Not Establish a Sequence

The Richmond Fed’s February 2026 review reports that BNPL users tend to carry higher balances on other unsecured credit products than nonusers. It also states that the relationship may not be causal. More borrowing could precede BNPL use, BNPL could affect other borrowing, or both could reflect conditions not identified by the data.

That evidence does not show that the same people appear in FHA mortgage delinquency, credit card delinquency, youth unemployment, or other American Distress Index inputs. It does not establish that a household uses BNPL only after exhausting savings, or that BNPL causes a later missed payment.

Reporting Can Change Without One Automatic Outcome

Some pay-in-four providers have begun reporting at least some loans, while others have not. The Richmond Fed describes broader adoption as gradual. If a loan reaches a credit bureau, the New York Fed says it may appear in the “other” category of household debt.

More reporting could make both successful repayment and missed payments visible. The effect on a person’s credit score depends on what a provider reports, which bureau receives it, how a scoring model treats it, and how a lender uses that score. The available evidence does not support an automatic score decline or one known jump in reported debt.

What This Means for the ADI

American Distress Index inputs describe different units and populations. A credit-file coverage gap does not tell us how far the Delinquency or Debt Burden domains are off, and it does not turn an annual BNPL volume estimate into missing household debt.

BNPL belongs in the distress picture as a measurement question: which loans are reported, which balances are observed, and which borrowers can be linked across products. Until one dataset answers those questions together, the honest result is a known gap with an unknown national balance.

What to Watch

Future New York Fed methodology notes can show whether reported BNPL is becoming more visible in the “other” category. CFPB lender collections can show annual loan activity within their stated samples. A national outstanding-balance estimate would require provider-level balances measured on the same date without double-counting borrowers or mixing pay-in-four loans with longer-term installment credit.

For the broader debt picture, see Consumer Debt Statistics and Household Debt Statistics. Read every comparison by its source, period, unit, and population before treating two figures as one measure.