U.S. Household Debt in Q2 2026: Credit Cards and Delinquency Measures

U.S. household debt stood at $18.771 trillions in Q2 2026, according to the Federal Reserve Bank of New York. That was $13 billion lower than in Q1 2026, a 0.1% quarterly decline, but $383 billion higher than in Q2 2025.

The quarterly decline was not broad based. The New York Fed reported that non-housing debt rose by $48 billion, or 0.9%, from Q1, while reported mortgage balances fell by $74 billion. The mortgage move needs an important reporting caveat from the New York Fed, discussed below.

Credit-card balances rose by $21 billion during Q2 to $1.263 trillion and were $54 billion higher than a year earlier. At the same time, the annualized flow of credit-card balances newly entering serious delinquency was 6.97% in Q2 2026, compared with 6.93% in Q2 2025.

Those figures describe different parts of the household-credit picture. They are useful together, but only if balances, stock delinquency and new delinquency transitions are kept separate.

Total household debt edged lower, but the composition matters

The New York Fed’s Q2 release gives the following balance levels and changes:

Debt type	Q2 2026 balance	Change from Q1 2026	Change from Q2 2025
Mortgage	$13.117 trillion	-$74 billion	+$182 billion
HELOC	$459 billion	+$13 billion	+$48 billion
Student loan	$1.651 trillion	-$7 billion	+$13 billion
Auto loan	$1.713 trillion	+$28 billion	+$58 billion
Credit card	$1.263 trillion	+$21 billion	+$54 billion
Other	$568 billion	+$6 billion	+$28 billion
Total	$18.771 trillion	-$13 billion	+$383 billion

The $13 billion decline in total debt should not be read as broad-based household deleveraging or as a decline across debt categories. The reported total is a debt-stock measure, not a measure of new borrowing activity.

Mortgage balances shown on consumer credit reports fell by $74 billion. The Quarterly Report attributes most of that reported decline to a “servicer transfer gap” in mortgage reporting and says balances otherwise would have “stayed flat.”

That distinction matters because mortgages are the largest component of household debt. Separately, the report says non-housing balances rose by $48 billion. Auto loans, credit cards and the “other” category all increased during Q2, while student-loan balances declined slightly.

The figures are nominal dollar balances. They are not adjusted for inflation, so changes over longer periods should not automatically be read as equivalent changes in real debt burden.

Credit-card balances increased by $21 billion

Credit-card balances reached $1.263 trillion in Q2 2026, up $21 billion, or 1.7%, from Q1. Compared with Q2 2025, balances were $54 billion higher.

Credit cards were among the categories with positive quarterly balance growth in the report. Auto-loan balances also rose, increasing $28 billion to $1.713 trillion. The New York Fed’s “other” category, which includes retail cards and consumer finance loans, increased $6 billion to $568 billion.

HELOC balances, which are housing-related rather than part of the report’s non-housing total, rose $13 billion to $459 billion. The report says this was the 17th consecutive quarterly increase in HELOC balances.

Student-loan balances moved in the opposite direction, declining $7 billion to $1.651 trillion during the quarter.

The useful point is not that every form of debt moved in the same direction. They did not. The Q2 data show a small decline in reported aggregate debt alongside increases in several non-mortgage categories.

The aggregate stock delinquency measure was slightly lower

At the end of June, 4.7% of outstanding debt was in some stage of delinquency, according to the Quarterly Report. That was 0.1 percentage points lower than in Q1 2026.

This is a stock measure. It describes the share of outstanding balances that were delinquent at the end of the quarter.

The report also publishes transition measures that ask a different question. For serious delinquency, the published rates are annualized using a four-quarter moving sum and relate newly serious-delinquent balances to balances that were not seriously delinquent in the previous quarter.

For serious delinquency, defined here as 90 or more days late, the annualized transition rates for several categories were close to their levels a year earlier:

Debt type	Q2 2025 flow into serious delinquency	Q2 2026 flow into serious delinquency	Change
Mortgage	1.29%	1.52%	+0.23 percentage points
HELOC	1.15%	1.15%	0.00 percentage points
Auto loan	2.93%	3.00%	+0.07 percentage points
Credit card	6.93%	6.97%	+0.04 percentage points
Other	5.42%	5.19%	-0.23 percentage points

For credit cards, the year-over-year change in the serious-delinquency transition rate was small: 6.97% in Q2 2026 versus 6.93% in Q2 2025.

That does not mean every credit-card delinquency measure was unchanged. It means that this particular measure, the annualized flow of balances newly becoming at least 90 days late, was nearly unchanged from a year earlier.

Two credit-card delinquency measures tell different parts of the story

The New York Fed report includes both a stock measure of credit-card delinquency and a flow measure of new serious-delinquency transitions.

The stock measure is the share of outstanding credit-card balances that are currently reported as 90 or more days delinquent. In the report’s underlying data, that measure was 12.92% in Q2 2026, down slightly from 13.12% in Q1 2026. It was 7.59% in Q3 2022.

The flow measure tracks balances that newly became at least 90 days late, relative to balances that were current or less than 90 days late in the previous quarter. The New York Fed annualizes this series using a four-quarter moving sum. That measure was 6.97% in Q2 2026, compared with 7.10% in Q1 2026 and 6.93% in Q2 2025.

These percentages should not be treated as competing versions of the same number. They use different definitions and denominators.

The companion Liberty Street Economics analysis by New York Fed researchers discusses why the stock measure can remain elevated even when the flow of new serious delinquencies is comparatively steady. The authors say the stock measure kept rising as “charged-off debts accumulated” and also point to lenders reporting those balances for “longer than they used to.”

The authors’ characterization of the flow measure is also explicitly their interpretation. They describe the pace of credit-card delinquency as elevated but “largely stable since 2024.”

For readers, the practical lesson is simpler: a stock measure tells you how much reported debt is already in a delinquent state, while a flow measure tells you how much debt is newly moving into that state. One can move differently from the other without either measure being incorrect.

Student-loan delinquency needs an extra caution

Student-loan balances were $1.651 trillion in Q2 2026, down $7 billion from Q1 and up $13 billion from Q2 2025.

The report says 10.6% of student-loan balances were 90 or more days past due in Q2, up from 10.3% in Q1. At the same time, the annualized flow into serious delinquency was 7.83% in Q2 2026, compared with 12.88% in Q2 2025.

Those figures are different measures, and the New York Fed adds a further warning. Its Q2 release points to the “re-reporting of defaulted student debt” as “causing some distortions” in student-loan delinquency measures.

For that reason, the student-loan delinquency figures are better treated cautiously rather than used as a simple signal that repayment conditions either improved or worsened by a particular amount.

What this means for debt planning

National household-debt data are useful for understanding the broader credit environment, but they do not determine what any one household should do.

The Q2 report shows that credit-card and auto-loan balances increased, non-housing debt rose overall, and the credit-card serious-delinquency transition rate was almost unchanged from a year earlier. It also shows why a small movement in total household debt can give an incomplete picture if the composition and reporting caveats are ignored.

For an individual debt plan, the relevant inputs are much more specific: the balances actually owed, interest rates, minimum payments, due dates, income, essential expenses and the cash available each month.

Aggregate statistics can provide context. They cannot tell a particular household whether an extra payment is affordable, which account should be prioritized, or how much room exists in next month’s budget. Those questions depend on the household’s own numbers.

Methodology and source note

The Quarterly Report on Household Debt and Credit is based on the New York Fed Consumer Credit Panel. The New York Fed describes the panel as an anonymized, nationally representative sample drawn from Equifax credit-report data.

The report’s data dictionary says the panel is built from a 5% random sample of individuals with a Social Security number and a credit report, together with people living at the same addresses for household-level analysis. The database includes about 44 million individuals in each quarter.

For most report graphs, the New York Fed says it uses a 0.1% random subsample of individuals with credit reports. Student-loan graphs use a 1% sample, while new foreclosure and bankruptcy rates use the full 5% sample.

These are credit-report-based debt measures. The report excludes some categories of trades and accounts that are not currently reported, so the totals should not be read as a complete inventory of every financial obligation held by U.S. households.

The figures in this article are nominal and come from the New York Fed’s Q2 2026 Quarterly Report, its August 11 release and the report’s official underlying data. The Liberty Street Economics companion article is used only for clearly attributed explanation of the credit-card delinquency measures.

Sources

1. Federal Reserve Bank of New York - Quarterly Report on Household Debt and Credit, 2026 Q2
2. Federal Reserve Bank of New York - Household Debt Balances Decreased Slightly; Credit Card Delinquency Transition Rates Remained Steady - August 11, 2026
3. Liberty Street Economics - How Distressed Are Consumers? Reconciling Diverging Credit Card Delinquency Measures - August 11, 2026
4. Federal Reserve Bank of New York - Official Q2 2026 Household Debt and Credit data workbook

Sources are provided so readers can review the public data and statements behind this article. MyDebtLens articles are educational only and are not financial advice.