Total household debt decreased $13 billion (0.1%) in the second quarter of 2026 to $18.77 trillion, according to the New York Fed’s latest Quarterly Report on Household Debt and Credit. Student loan balances fell $7 billion to $1.65 trillion – one of only two categories to shrink, alongside mortgages (down $74 billion).

Credit card balances rose $21 billion to $1.26 trillion, keeping total student loan debt as the second-largest consumer debt category behind housing.

The more notable number is buried in the delinquency tables: the share of student loan balances newly transitioning into serious (90+ day) delinquency fell to 7.83% in Q2 2026, down from 12.88% a year earlier. The Fed cautions that re-reporting of defaulted student debt is still distorting the student loan figures.

Did Resumed Payments Help The Total Balance Fall?

Likely yes, though the Fed doesn’t break out the reasons why a balance fell.

Second-quarter dips are a recurring seasonal pattern in this data because few new federal loans are disbursed between spring and fall semesters. An average of $90-100 billion in new debt happens each year, but those happen in August/September, and January. Then, current borrowers repay their loans all year. With no new originations in Spring generally, debt balances can fall overall.

But the repayment system has also restarted in stages over the past year: interest resumed for roughly 8 million SAVE borrowers in August 2025, and the SAVE forbearance end is forcing about 7 million borrowers to pick new repayment plans and start paying again.

More borrowers making real payments pulls balances down, even with balances still $13 billion higher than a year ago. 

Tracking Student Loan Default

Recent Associated Press reporting put “record” defaults at roughly 9.5 million borrowers and $233.3 billion.

We found this reporting to be misleading when you read the fine print: AP created that number by adding two Education Department delinquency buckets (borrowers 271 to 360 days behind and borrowers more than 360 days behind) and labeling them “defaulted borrowers.”

The Education Department data behind that count runs only through March 2026 and no newer official default numbers have been published since. 

This claim rests on an assumption that borrowers who were delinquent in March stayed there and didn’t make any effort to get back on track. That’s an extrapolation presented as a fact, and is misleading.

The new Fed data is the first major independent read since March, and it doesn’t necessarily support that assumption. New borrowers falling into serious delinquency fell to 7.83% from 12.88% a year earlier, and the Fed described transition rates across most products as steady. That’s a sign that the post-payment restart default wave cresting, not a system accelerating toward more defaults. 

Federal loans do technically hit default at 270 days of missed payments, but student loan borrowers specifically haven’t been transferred to collections yet and can still get current. With reports of15% wage garnishment on defaulted loans resuming, and tax offsets coming back, many borrowers may be taking that action. 

How This Connects

None of this means student loan borrowers are fine.

Credit scores dropped sharply when delinquency reporting resumed, and the consequences of default are returning: Treasury now handles collections for millions of defaulted borrowers, including tax refund offsets and garnishment.

But the distinction between real defaults and headlines matter, because a borrower who is 300 days behind still has time to act before collections start, and policy debates should run on actual data, not assumptions.

Student loan balances fell, new serious delinquencies dropped by nearly 5%, and the official default count is now five months stale.

Until the Education Department publishes fresh data, any “record default” number is an estimate built on assumptions and reporting that skips that caveat isn’t informing borrowers, it’s scaring them.

We track the verified numbers on our student loan delinquency coverage as they publish.

What’s Next

• Federal Student Aid’s next quarterly portfolio release will deliver the first official default count since March ad we will really see which way defaults are heading.
• The Fed’s Q3 report arrives in November and will show whether balances keep falling as interest accrues for former SAVE borrowers now in IBR or standard plans.
• Watch whether ramped-up garnishment and offsets push loan rehabilitation and consolidation activity higher as defaulted borrowers look for a way out.
• Longer term, existing income-driven plans wind down by July 2028 under the new rules, which will likely keep these numbers in flux for a while.

Don’t Miss These Other Stories: