U.S. Household Debt Statistics 2026: The $18.8 Trillion Balance Sheet

U.S. household debt reached $18.8 trillion at the end of the second quarter of 2026, a $13 billion, or 0.1%, decline from the prior quarter. That dip is not clean evidence of household deleveraging: mortgages fell $74 billion partly because of a reporting gap, while non-housing debt grew $48 billion.

What the balance sheet contains

Mortgages remain the dominant part of American household debt. Mortgage balances stood at $13.1 trillion, far larger than any other category listed in the release. That makes the mortgage number important, but it also makes changes in mortgage reporting especially capable of moving the headline total.

Home equity lines of credit, or HELOCs, reached $459 billion. The balance has increased for 17 consecutive quarters and now stands $142 billion above its low in early 2022.

Credit cards and auto loans are smaller than mortgages but more closely tied to everyday spending and transportation costs. Credit card balances were $1.26 trillion, while auto loan balances reached $1.71 trillion.

The table below puts selected balances against the full household-debt total. These are shares of outstanding debt, not shares of household income, wealth, or monthly cash flow.

The listed categories do not represent every balance in the release. Student-loan balances are discussed separately because the supplied facts do not provide their dollar balance.

What changed during the quarter

The most important movement was outside housing. Non-housing debt grew $48 billion, or 0.9%. Credit cards increased $21 billion, or 1.7%. Auto loans also rose $28 billion, or 1.7%.

HELOC balances rose $13 billion. Other balances, including retail cards and consumer finance loans, rose $6 billion to $568 billion. Student-loan balances declined 0.4%.

Credit cards plus auto loans totaled $2.97 trillion, equal to 15.8% of total household debt. This PlainReads calculation shows why the composition matters: these categories are much smaller than mortgages, but their balances changed more visibly during the quarter.

Why the headline decline needs context

Mortgage balances declined $74 billion. The New York Fed says a temporary reporting gap caused by a mortgage-servicing transfer contributed to that decline. A servicing transfer can change how an account appears in reported data without representing an equivalent change in the borrower’s underlying obligation.

That caveat changes the interpretation of the total. Household debt fell $13 billion, but non-housing balances grew $48 billion. The mortgage reporting issue therefore makes the headline balance look weaker than the underlying mix of borrowing suggests.

The HELOC trend adds another layer. HELOC balances are now $142 billion above their low in early 2022. Using the stated balance and increase, the implied low was about $317 billion, making the increase 44.8%. This is a PlainReads calculation, not a separate reported figure.

How to interpret the mix

Debt composition matters more than the headline total alone. Mortgages dominate the stock, so their movements can overwhelm smaller changes elsewhere. Yet non-housing borrowing drove the quarter’s growth pattern. A household can therefore appear less indebted in the aggregate while carrying more credit-card, auto-loan, or HELOC debt.

That does not mean non-housing borrowing is automatically harmful. Debt used to buy a home, finance transportation, smooth a temporary cash-flow gap, or fund a purchase can have different consequences. The balance itself does not reveal the interest rate, repayment schedule, income support, or purpose of the borrowing.

A useful next step is to separate what a household owns from what it owes with assets vs liabilities. For households managing several payment obligations, a zero-based budget can make the monthly consequences of these balances easier to see. Over longer periods, the Compounding Flywheel explains why repayment, saving, and reinvestment can reinforce one another.

Delinquency is not the same as debt growth

At the end of the quarter, 4.7% of outstanding debt was in some stage of delinquency, down 0.1 percentage point from the prior quarter. Serious-delinquency transition rates were unchanged.

The denominator matters. This delinquency rate applies to outstanding debt, not to households, borrowers, new loans, or the amount borrowed during the quarter. A stable rate can coexist with rising balances, falling balances, or a changing mix of loan types.

Student loans require special caution. The release says that re-reporting defaulted student debt distorts delinquency rates. That means a change in reported student-loan delinquency may partly reflect how defaulted debt re-enters the data rather than a straightforward change in borrower behavior.

Correlation also does not establish causation. The simultaneous rise in credit-card and auto-loan balances does not prove that one caused the other, nor does it prove that mortgage-servicing activity caused households to borrow elsewhere. The figures show movements occurring in the same reporting period; they do not identify the personal or economic causes behind every movement.

Methodology and limitations

The release is an aggregate snapshot of household credit balances and delinquency status at the end of the second quarter of 2026. It was released in August 2026. The comparison is with the prior quarter, not with the same period in the prior year.

This is not a survey of household opinions, financial confidence, or spending plans. The supplied facts do not provide a household count, income distribution, wealth distribution, or borrower-level repayment capacity. No household average or per-person figure should therefore be inferred from the aggregate balance.

Seasonality also limits a single quarter’s meaning. Borrowing and repayments can vary during the calendar year, and a quarter-to-quarter change can reflect timing as well as lasting behavior. The supplied release does not provide a seasonal explanation for each category, so the safest reading is to treat these figures as a current snapshot and compare future releases before declaring a trend.

Reporting changes are especially important here. The mortgage-servicing transfer gap affects the mortgage comparison, while student-loan re-reporting affects delinquency interpretation. Later releases may revise historical comparisons or provide updated context. When that happens, the latest New York Fed release should take priority over this snapshot.

Sources

All reported figures and PlainReads calculations in this article use the Federal Reserve Bank of New York Household Debt and Credit report for the second quarter of 2026.

Frequently asked questions

Is household debt falling?

The reported total declined $13 billion, or 0.1%, from the prior quarter. That does not necessarily mean households broadly reduced their obligations because a mortgage-servicing transfer reporting gap contributed to the $74 billion mortgage decline.

Which debt category is largest?

Mortgages are largest at $13.1 trillion, representing 69.7% of the total by PlainReads calculation. Their size means mortgage reporting changes can strongly affect the headline household-debt balance.

Why are HELOCs important?

HELOC balances reached $459 billion, after 17 consecutive quarters of increases. They are $142 billion above their low in early 2022, an increase of 44.8% by PlainReads calculation.

Does the delinquency rate show household financial health?

It provides a useful credit-performance indicator, but not a complete measure of financial health. 4.7% of outstanding debt was delinquent, yet the rate does not show household income, savings, assets, or the effect of student-loan re-reporting.